diff --git a/data/xml/2020.acl.xml b/data/xml/2020.acl.xml
index 597486a33f..a832fcc617 100644
--- a/data/xml/2020.acl.xml
+++ b/data/xml/2020.acl.xml
@@ -9326,6 +9326,7 @@
       <video href="http://slideslive.com/38928734"/>
       <bibkey>peinelt-etal-2020-tbert</bibkey>
       <pwccode url="https://github.com/wuningxi/tBERT" additional="false">wuningxi/tBERT</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/100doh">100DOH</pwcdataset>
     </paper>
     <paper id="631">
       <title>Conditional Augmentation for Aspect Term Extraction via Masked Sequence-to-Sequence Generation</title>
diff --git a/data/xml/2020.coling.xml b/data/xml/2020.coling.xml
index bf03379de7..a7b6ecb0c2 100644
--- a/data/xml/2020.coling.xml
+++ b/data/xml/2020.coling.xml
@@ -6474,7 +6474,7 @@
       <url hash="d5041c46">2020.coling-main.488</url>
       <doi>10.18653/v1/2020.coling-main.488</doi>
       <bibkey>schick-etal-2020-automatically</bibkey>
-      <pwccode url="https://github.com/timoschick/pet" additional="false">timoschick/pet</pwccode>
+      <pwccode url="https://github.com/timoschick/pet" additional="true">timoschick/pet</pwccode>
     </paper>
     <paper id="489">
       <title>Knowledge Base Embedding By Cooperative Knowledge Distillation</title>
diff --git a/data/xml/2021.acl.xml b/data/xml/2021.acl.xml
index f848ddfc35..a6fec33538 100644
--- a/data/xml/2021.acl.xml
+++ b/data/xml/2021.acl.xml
@@ -31,8 +31,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="6703d0d5">2021.acl-long.1.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.1</doi>
       <bibkey>beck-etal-2021-investigating</bibkey>
-      <pwccode url="https://github.com/UKPLab/acl2021-label-suggestions-german-covid19" additional="false">UKPLab/acl2021-label-suggestions-german-covid19</pwccode>
       <video href="2021.acl-long.1.mp4"/>
+      <pwccode url="https://github.com/UKPLab/acl2021-label-suggestions-german-covid19" additional="false">UKPLab/acl2021-label-suggestions-german-covid19</pwccode>
     </paper>
     <paper id="2">
       <title><fixed-case>H</fixed-case>ow Did This Get Funded?! <fixed-case>A</fixed-case>utomatically Identifying Quirky Scientific Achievements</title>
@@ -44,8 +44,8 @@
       <url hash="00007d04">2021.acl-long.2</url>
       <doi>10.18653/v1/2021.acl-long.2</doi>
       <bibkey>shani-etal-2021-get</bibkey>
-      <pwccode url="https://github.com/nadavborenstein/Iggy" additional="false">nadavborenstein/Iggy</pwccode>
       <video href="2021.acl-long.2.mp4"/>
+      <pwccode url="https://github.com/nadavborenstein/Iggy" additional="false">nadavborenstein/Iggy</pwccode>
     </paper>
     <paper id="3">
       <title>Engage the Public: Poll Question Generation for Social Media Posts</title>
@@ -60,8 +60,8 @@
       <url hash="7ae7115a">2021.acl-long.3</url>
       <doi>10.18653/v1/2021.acl-long.3</doi>
       <bibkey>lu-etal-2021-engage</bibkey>
-      <pwccode url="https://github.com/polyusmart/poll-question-generation" additional="false">polyusmart/poll-question-generation</pwccode>
       <video href="2021.acl-long.3.mp4"/>
+      <pwccode url="https://github.com/polyusmart/poll-question-generation" additional="false">polyusmart/poll-question-generation</pwccode>
     </paper>
     <paper id="4">
       <title><fixed-case>H</fixed-case>ate<fixed-case>C</fixed-case>heck: Functional Tests for Hate Speech Detection Models</title>
@@ -76,8 +76,8 @@
       <url hash="1349cfa3">2021.acl-long.4</url>
       <doi>10.18653/v1/2021.acl-long.4</doi>
       <bibkey>rottger-etal-2021-hatecheck</bibkey>
-      <pwccode url="https://github.com/paul-rottger/hate-functional-tests" additional="true">paul-rottger/hate-functional-tests</pwccode>
       <video href="2021.acl-long.4.mp4"/>
+      <pwccode url="https://github.com/paul-rottger/hate-functional-tests" additional="true">paul-rottger/hate-functional-tests</pwccode>
     </paper>
     <paper id="5">
       <title>Unified Dual-view Cognitive Model for Interpretable Claim Verification</title>
@@ -91,8 +91,8 @@
       <url hash="40de9c9b">2021.acl-long.5</url>
       <doi>10.18653/v1/2021.acl-long.5</doi>
       <bibkey>wu-etal-2021-unified</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <video href="2021.acl-long.5.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
     </paper>
     <paper id="6">
       <title><fixed-case>D</fixed-case>eep<fixed-case>R</fixed-case>apper: Neural Rap Generation with Rhyme and Rhythm Modeling</title>
@@ -124,8 +124,8 @@
       <url hash="79e4141f">2021.acl-long.7</url>
       <doi>10.18653/v1/2021.acl-long.7</doi>
       <bibkey>ao-etal-2021-pens</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
       <video href="2021.acl-long.7.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
     </paper>
     <paper id="8">
       <title>Enhancing Content Preservation in Text Style Transfer Using Reverse Attention and Conditional Layer Normalization</title>
@@ -138,8 +138,8 @@
       <url hash="cca9d69f">2021.acl-long.8</url>
       <doi>10.18653/v1/2021.acl-long.8</doi>
       <bibkey>lee-etal-2021-enhancing</bibkey>
-      <pwccode url="https://github.com/MovingKyu/RACoLN" additional="false">MovingKyu/RACoLN</pwccode>
       <video href="2021.acl-long.8.mp4"/>
+      <pwccode url="https://github.com/MovingKyu/RACoLN" additional="false">MovingKyu/RACoLN</pwccode>
     </paper>
     <paper id="9">
       <title>Mention Flags (<fixed-case>MF</fixed-case>): Constraining Transformer-based Text Generators</title>
@@ -154,11 +154,11 @@
       <attachment type="OptionalSupplementaryMaterial" hash="580ee7ed">2021.acl-long.9.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.9</doi>
       <bibkey>wang-etal-2021-mention</bibkey>
+      <video href="2021.acl-long.9.mp4"/>
       <pwccode url="https://github.com/garyyufei/acl2021mf" additional="false">garyyufei/acl2021mf</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/commongen">CommonGen</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/nocaps">nocaps</pwcdataset>
-      <video href="2021.acl-long.9.mp4"/>
     </paper>
     <paper id="10">
       <title>Generalising Multilingual Concept-to-Text <fixed-case>NLG</fixed-case> with Language Agnostic Delexicalisation</title>
@@ -169,8 +169,8 @@
       <url hash="0744ef0f">2021.acl-long.10</url>
       <doi>10.18653/v1/2021.acl-long.10</doi>
       <bibkey>zhou-lampouras-2021-generalising</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/crosswoz">CrossWOZ</pwcdataset>
       <video href="2021.acl-long.10.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/crosswoz">CrossWOZ</pwcdataset>
     </paper>
     <paper id="11">
       <title>Conversations Are Not Flat: Modeling the Dynamic Information Flow across Dialogue Utterances</title>
@@ -184,10 +184,10 @@
       <url hash="b843f9fc">2021.acl-long.11</url>
       <doi>10.18653/v1/2021.acl-long.11</doi>
       <bibkey>li-etal-2021-conversations</bibkey>
+      <video href="2021.acl-long.11.mp4"/>
       <pwccode url="https://github.com/ictnlp/DialoFlow" additional="false">ictnlp/DialoFlow</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fed">FED</pwcdataset>
-      <video href="2021.acl-long.11.mp4"/>
     </paper>
     <paper id="12">
       <title>Dual Slot Selector via Local Reliability Verification for Dialogue State Tracking</title>
@@ -200,9 +200,9 @@
       <url hash="edf4742b">2021.acl-long.12</url>
       <doi>10.18653/v1/2021.acl-long.12</doi>
       <bibkey>guo-etal-2021-dual</bibkey>
+      <video href="2021.acl-long.12.mp4"/>
       <pwccode url="https://github.com/guojinyu88/DSSDST" additional="false">guojinyu88/DSSDST</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
-      <video href="2021.acl-long.12.mp4"/>
     </paper>
     <paper id="13">
       <title>Transferable Dialogue Systems and User Simulators</title>
@@ -215,8 +215,8 @@
       <url hash="b6dad635">2021.acl-long.13</url>
       <doi>10.18653/v1/2021.acl-long.13</doi>
       <bibkey>tseng-etal-2021-transferable</bibkey>
-      <pwccode url="https://github.com/andy194673/joust" additional="false">andy194673/joust</pwccode>
       <video href="2021.acl-long.13.mp4"/>
+      <pwccode url="https://github.com/andy194673/joust" additional="false">andy194673/joust</pwccode>
     </paper>
     <paper id="14">
       <title><fixed-case>B</fixed-case>o<fixed-case>B</fixed-case>: <fixed-case>BERT</fixed-case> Over <fixed-case>BERT</fixed-case> for Training Persona-based Dialogue Models from Limited Personalized Data</title>
@@ -230,11 +230,11 @@
       <url hash="0e16445d">2021.acl-long.14</url>
       <doi>10.18653/v1/2021.acl-long.14</doi>
       <bibkey>song-etal-2021-bob</bibkey>
+      <video href="2021.acl-long.14.mp4"/>
       <pwccode url="https://github.com/songhaoyu/BoB" additional="false">songhaoyu/BoB</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/convai2">ConvAI2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/personaldialog">PersonalDialog</pwcdataset>
-      <video href="2021.acl-long.14.mp4"/>
     </paper>
     <paper id="15">
       <title><fixed-case>GL</fixed-case>-<fixed-case>GIN</fixed-case>: Fast and Accurate Non-Autoregressive Model for Joint Multiple Intent Detection and Slot Filling</title>
@@ -249,10 +249,10 @@
       <url hash="20971636">2021.acl-long.15</url>
       <doi>10.18653/v1/2021.acl-long.15</doi>
       <bibkey>qin-etal-2021-gl</bibkey>
+      <video href="2021.acl-long.15.mp4"/>
       <pwccode url="https://github.com/yizhen20133868/GL-GIN" additional="false">yizhen20133868/GL-GIN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.acl-long.15.mp4"/>
     </paper>
     <paper id="16">
       <title>Accelerating <fixed-case>BERT</fixed-case> Inference for Sequence Labeling via Early-Exit</title>
@@ -267,11 +267,11 @@
       <url hash="3cf4a8fa">2021.acl-long.16</url>
       <doi>10.18653/v1/2021.acl-long.16</doi>
       <bibkey>li-etal-2021-accelerating</bibkey>
+      <video href="2021.acl-long.16.mp4"/>
       <pwccode url="https://github.com/LeeSureman/Sequence-Labeling-Early-Exit" additional="false">LeeSureman/Sequence-Labeling-Early-Exit</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clue">CLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.acl-long.16.mp4"/>
     </paper>
     <paper id="17">
       <title>Modularized Interaction Network for Named Entity Recognition</title>
@@ -288,9 +288,9 @@
       <url hash="8afc223b">2021.acl-long.17</url>
       <doi>10.18653/v1/2021.acl-long.17</doi>
       <bibkey>li-etal-2021-modularized</bibkey>
+      <video href="2021.acl-long.17.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wnut-2017-emerging-and-rare-entity">WNUT 2017</pwcdataset>
-      <video href="2021.acl-long.17.mp4"/>
     </paper>
     <paper id="18">
       <title>Capturing Event Argument Interaction via A Bi-Directional Entity-Level Recurrent Decoder</title>
@@ -320,9 +320,9 @@
       <url hash="4d06c1aa">2021.acl-long.19</url>
       <doi>10.18653/v1/2021.acl-long.19</doi>
       <bibkey>wang-etal-2021-unire</bibkey>
+      <video href="2021.acl-long.19.mp4"/>
       <pwccode url="https://github.com/Receiling/UniRE" additional="false">Receiling/UniRE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
-      <video href="2021.acl-long.19.mp4"/>
     </paper>
     <paper id="20">
       <title>Refining Sample Embeddings with Relation Prototypes to Enhance Continual Relation Extraction</title>
@@ -339,9 +339,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="4b0ce532">2021.acl-long.20.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.20</doi>
       <bibkey>cui-etal-2021-refining</bibkey>
+      <video href="2021.acl-long.20.mp4"/>
       <pwccode url="https://github.com/fd2014cl/rp-cre" additional="false">fd2014cl/rp-cre</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
-      <video href="2021.acl-long.20.mp4"/>
     </paper>
     <paper id="21">
       <title>Contrastive Learning for Many-to-many Multilingual Neural Machine Translation</title>
@@ -354,9 +354,9 @@
       <url hash="359e1df2">2021.acl-long.21</url>
       <doi>10.18653/v1/2021.acl-long.21</doi>
       <bibkey>pan-etal-2021-contrastive</bibkey>
+      <video href="2021.acl-long.21.mp4"/>
       <pwccode url="https://github.com/PANXiao1994/mCOLT" additional="true">PANXiao1994/mCOLT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/opus-100">OPUS-100</pwcdataset>
-      <video href="2021.acl-long.21.mp4"/>
     </paper>
     <paper id="22">
       <title>Understanding the Properties of Minimum <fixed-case>B</fixed-case>ayes Risk Decoding in Neural Machine Translation</title>
@@ -367,8 +367,8 @@
       <url hash="58ea5782">2021.acl-long.22</url>
       <doi>10.18653/v1/2021.acl-long.22</doi>
       <bibkey>muller-sennrich-2021-understanding</bibkey>
-      <pwccode url="https://github.com/ZurichNLP/understanding-mbr" additional="false">ZurichNLP/understanding-mbr</pwccode>
       <video href="2021.acl-long.22.mp4"/>
+      <pwccode url="https://github.com/ZurichNLP/understanding-mbr" additional="false">ZurichNLP/understanding-mbr</pwccode>
     </paper>
     <paper id="23">
       <title>Multi-Head Highly Parallelized <fixed-case>LSTM</fixed-case> Decoder for Neural Machine Translation</title>
@@ -406,8 +406,8 @@
       <url hash="37c63029">2021.acl-long.25</url>
       <doi>10.18653/v1/2021.acl-long.25</doi>
       <bibkey>lin-etal-2021-learning</bibkey>
-      <pwccode url="https://github.com/NLP-Playground/LaSS" additional="false">NLP-Playground/LaSS</pwccode>
       <video href="2021.acl-long.25.mp4"/>
+      <pwccode url="https://github.com/NLP-Playground/LaSS" additional="false">NLP-Playground/LaSS</pwccode>
     </paper>
     <paper id="26">
       <title>Exploring the Efficacy of Automatically Generated Counterfactuals for Sentiment Analysis</title>
@@ -422,9 +422,9 @@
       <url hash="4f703108">2021.acl-long.26</url>
       <doi>10.18653/v1/2021.acl-long.26</doi>
       <bibkey>yang-etal-2021-exploring</bibkey>
+      <video href="2021.acl-long.26.mp4"/>
       <pwccode url="https://github.com/lijiazheng99/Counterfactuals-for-Sentiment-Analysis" additional="false">lijiazheng99/Counterfactuals-for-Sentiment-Analysis</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.acl-long.26.mp4"/>
     </paper>
     <paper id="27">
       <title>Bridge-Based Active Domain Adaptation for Aspect Term Extraction</title>
@@ -435,8 +435,8 @@
       <url hash="cee14744">2021.acl-long.27</url>
       <doi>10.18653/v1/2021.acl-long.27</doi>
       <bibkey>chen-qian-2021-bridge</bibkey>
-      <pwccode url="https://github.com/nlpwm-whu/bridge" additional="false">nlpwm-whu/bridge</pwccode>
       <video href="2021.acl-long.27.mp4"/>
+      <pwccode url="https://github.com/nlpwm-whu/bridge" additional="false">nlpwm-whu/bridge</pwccode>
     </paper>
     <paper id="28">
       <title>Multimodal Sentiment Detection Based on Multi-channel Graph Neural Networks</title>
@@ -449,8 +449,8 @@
       <url hash="a4b04a21">2021.acl-long.28</url>
       <doi>10.18653/v1/2021.acl-long.28</doi>
       <bibkey>yang-etal-2021-multimodal</bibkey>
-      <pwccode url="https://github.com/yangxiaocui1215/mgnns" additional="false">yangxiaocui1215/mgnns</pwccode>
       <video href="2021.acl-long.28.mp4"/>
+      <pwccode url="https://github.com/yangxiaocui1215/mgnns" additional="false">yangxiaocui1215/mgnns</pwccode>
     </paper>
     <paper id="29">
       <title>Aspect-Category-Opinion-Sentiment Quadruple Extraction with Implicit Aspects and Opinions</title>
@@ -462,10 +462,10 @@
       <url hash="ef9db799">2021.acl-long.29</url>
       <doi>10.18653/v1/2021.acl-long.29</doi>
       <bibkey>cai-etal-2021-aspect</bibkey>
+      <video href="2021.acl-long.29.mp4"/>
       <pwccode url="https://github.com/nustm/acos" additional="false">nustm/acos</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/laptop-acos">Laptop-ACOS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/restaurant-acos">Restaurant-ACOS</pwcdataset>
-      <video href="2021.acl-long.29.mp4"/>
     </paper>
     <paper id="30">
       <title><fixed-case>PASS</fixed-case>: Perturb-and-Select Summarizer for Product Reviews</title>
@@ -533,8 +533,8 @@
       <url hash="7f6fb25e">2021.acl-long.34</url>
       <doi>10.18653/v1/2021.acl-long.34</doi>
       <bibkey>chen-etal-2021-training</bibkey>
-      <pwccode url="https://github.com/Chen-Wang-CUHK/Training-Free-and-Ref-Free-Summ-Evaluation" additional="false">Chen-Wang-CUHK/Training-Free-and-Ref-Free-Summ-Evaluation</pwccode>
       <video href="2021.acl-long.34.mp4"/>
+      <pwccode url="https://github.com/Chen-Wang-CUHK/Training-Free-and-Ref-Free-Summ-Evaluation" additional="false">Chen-Wang-CUHK/Training-Free-and-Ref-Free-Summ-Evaluation</pwccode>
     </paper>
     <paper id="35">
       <title><fixed-case>DESCGEN</fixed-case>: A Distantly Supervised Datasetfor Generating Entity Descriptions</title>
@@ -546,10 +546,10 @@
       <url hash="b033c710">2021.acl-long.35</url>
       <doi>10.18653/v1/2021.acl-long.35</doi>
       <bibkey>shi-etal-2021-descgen</bibkey>
+      <video href="2021.acl-long.35.mp4"/>
       <pwccode url="https://github.com/swj0419/DESCGEN" additional="false">swj0419/DESCGEN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/realnews">RealNews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisum">WikiSum</pwcdataset>
-      <video href="2021.acl-long.35.mp4"/>
     </paper>
     <paper id="36">
       <title>Introducing Orthogonal Constraint in Structural Probes</title>
@@ -560,9 +560,9 @@
       <url hash="13c2b4b0">2021.acl-long.36</url>
       <doi>10.18653/v1/2021.acl-long.36</doi>
       <bibkey>limisiewicz-marecek-2021-introducing</bibkey>
+      <video href="2021.acl-long.36.mp4"/>
       <pwccode url="https://github.com/Tom556/OrthogonalTransformerProbing" additional="false">Tom556/OrthogonalTransformerProbing</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/english-web-treebank">English Web Treebank</pwcdataset>
-      <video href="2021.acl-long.36.mp4"/>
     </paper>
     <paper id="37">
       <title>Hidden Killer: Invisible Textual Backdoor Attacks with Syntactic Trigger</title>
@@ -578,10 +578,10 @@
       <url hash="08238446">2021.acl-long.37</url>
       <doi>10.18653/v1/2021.acl-long.37</doi>
       <bibkey>qi-etal-2021-hidden</bibkey>
+      <video href="2021.acl-long.37.mp4"/>
       <pwccode url="https://github.com/thunlp/HiddenKiller" additional="true">thunlp/HiddenKiller</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/olid">OLID</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.37.mp4"/>
     </paper>
     <paper id="38">
       <title>Examining the Inductive Bias of Neural Language Models with Artificial Languages</title>
@@ -592,8 +592,8 @@
       <url hash="9c983b4e">2021.acl-long.38</url>
       <doi>10.18653/v1/2021.acl-long.38</doi>
       <bibkey>white-cotterell-2021-examining</bibkey>
-      <pwccode url="https://github.com/rycolab/artificial-languages" additional="false">rycolab/artificial-languages</pwccode>
       <video href="2021.acl-long.38.mp4"/>
+      <pwccode url="https://github.com/rycolab/artificial-languages" additional="false">rycolab/artificial-languages</pwccode>
     </paper>
     <paper id="39">
       <title>Explaining Contextualization in Language Models using Visual Analytics</title>
@@ -620,9 +620,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="c58dcd11">2021.acl-long.40.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-long.40</doi>
       <bibkey>chrysostomou-aletras-2021-improving</bibkey>
+      <video href="2021.acl-long.40.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.40.mp4"/>
     </paper>
     <paper id="41">
       <title>Generating Landmark Navigation Instructions from Maps as a Graph-to-Text Problem</title>
@@ -633,9 +633,9 @@
       <url hash="2dd74d19">2021.acl-long.41</url>
       <doi>10.18653/v1/2021.acl-long.41</doi>
       <bibkey>schumann-riezler-2021-generating</bibkey>
+      <video href="2021.acl-long.41.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/map2seq">map2seq</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/talk-the-walk">Talk the Walk</pwcdataset>
-      <video href="2021.acl-long.41.mp4"/>
     </paper>
     <paper id="42">
       <title><fixed-case>E</fixed-case>2<fixed-case>E</fixed-case>-<fixed-case>VLP</fixed-case>: End-to-End Vision-Language Pre-training Enhanced by Visual Learning</title>
@@ -651,11 +651,11 @@
       <url hash="f8ad6328">2021.acl-long.42</url>
       <doi>10.18653/v1/2021.acl-long.42</doi>
       <bibkey>xu-etal-2021-e2e</bibkey>
+      <video href="2021.acl-long.42.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.acl-long.42.mp4"/>
     </paper>
     <paper id="43">
       <title>Learning Relation Alignment for Calibrated Cross-modal Retrieval</title>
@@ -672,10 +672,10 @@
       <url hash="48c83122">2021.acl-long.43</url>
       <doi>10.18653/v1/2021.acl-long.43</doi>
       <bibkey>ren-etal-2021-learning</bibkey>
+      <video href="2021.acl-long.43.mp4"/>
       <pwccode url="https://github.com/lancopku/IAIS" additional="false">lancopku/IAIS</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
-      <video href="2021.acl-long.43.mp4"/>
     </paper>
     <paper id="44">
       <title><fixed-case>KM</fixed-case>-<fixed-case>BART</fixed-case>: Knowledge Enhanced Multimodal <fixed-case>BART</fixed-case> for Visual Commonsense Generation</title>
@@ -691,13 +691,13 @@
       <attachment type="OptionalSupplementaryMaterial" hash="7d96a20e">2021.acl-long.44.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-long.44</doi>
       <bibkey>xing-etal-2021-km</bibkey>
+      <video href="2021.acl-long.44.mp4"/>
       <pwccode url="https://github.com/FomalhautB/KM-BART-ACL" additional="false">FomalhautB/KM-BART-ACL</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.acl-long.44.mp4"/>
     </paper>
     <paper id="45">
       <title>Cascaded Head-colliding Attention</title>
@@ -709,10 +709,11 @@
       <url hash="b807d39c">2021.acl-long.45</url>
       <doi>10.18653/v1/2021.acl-long.45</doi>
       <bibkey>zheng-etal-2021-cascaded</bibkey>
+      <video href="2021.acl-long.45.mp4"/>
       <pwccode url="https://github.com/LZhengisme/CODA" additional="false">LZhengisme/CODA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.acl-long.45.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="46">
       <title>Structural Knowledge Distillation: Tractably Distilling Information for Structured Predictor</title>
@@ -730,12 +731,12 @@
       <url hash="846ab959">2021.acl-long.46</url>
       <doi>10.18653/v1/2021.acl-long.46</doi>
       <bibkey>wang-etal-2021-structural</bibkey>
+      <video href="2021.acl-long.46.mp4"/>
       <pwccode url="https://github.com/Alibaba-NLP/StructuralKD" additional="false">Alibaba-NLP/StructuralKD</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2002">CoNLL 2002</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikiann-1">WikiAnn</pwcdataset>
-      <video href="2021.acl-long.46.mp4"/>
     </paper>
     <paper id="47">
       <title>Parameter-efficient Multi-task Fine-tuning for Transformers via Shared Hypernetworks</title>
@@ -748,6 +749,7 @@
       <url hash="4ae58eba">2021.acl-long.47</url>
       <doi>10.18653/v1/2021.acl-long.47</doi>
       <bibkey>karimi-mahabadi-etal-2021-parameter</bibkey>
+      <video href="2021.acl-long.47.mp4"/>
       <pwccode url="https://github.com/rabeehk/hyperformer" additional="false">rabeehk/hyperformer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
@@ -756,7 +758,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/paws">PAWS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.acl-long.47.mp4"/>
     </paper>
     <paper id="48">
       <title><fixed-case>COSY</fixed-case>: <fixed-case>CO</fixed-case>unterfactual <fixed-case>SY</fixed-case>ntax for Cross-Lingual Understanding</title>
@@ -770,10 +771,10 @@
       <url hash="ca5b695d">2021.acl-long.48</url>
       <doi>10.18653/v1/2021.acl-long.48</doi>
       <bibkey>yu-etal-2021-cosy</bibkey>
+      <video href="2021.acl-long.48.mp4"/>
       <pwccode url="https://github.com/pluviophileyu/cosy" additional="false">pluviophileyu/cosy</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.acl-long.48.mp4"/>
     </paper>
     <paper id="49">
       <title><fixed-case>O</fixed-case>o<fixed-case>MM</fixed-case>ix: Out-of-manifold Regularization in Contextual Embedding Space for Text Classification</title>
@@ -786,8 +787,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="d860dc5b">2021.acl-long.49.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.49</doi>
       <bibkey>lee-etal-2021-oommix</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <video href="2021.acl-long.49.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
     </paper>
     <paper id="50">
       <title>Understanding and Countering Stereotypes: A Computational Approach to the Stereotype Content Model</title>
@@ -799,8 +800,8 @@
       <url hash="03e1352c">2021.acl-long.50</url>
       <doi>10.18653/v1/2021.acl-long.50</doi>
       <bibkey>fraser-etal-2021-understanding</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/stereoset">StereoSet</pwcdataset>
       <video href="2021.acl-long.50.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/stereoset">StereoSet</pwcdataset>
     </paper>
     <paper id="51">
       <title>Structurizing Misinformation Stories via Rationalizing Fact-Checks</title>
@@ -837,8 +838,8 @@
       <url hash="7d5850e7">2021.acl-long.53</url>
       <doi>10.18653/v1/2021.acl-long.53</doi>
       <bibkey>sahai-etal-2021-breaking</bibkey>
-      <pwccode url="https://github.com/sahaisaumya/informal_fallacies" additional="false">sahaisaumya/informal_fallacies</pwccode>
       <video href="2021.acl-long.53.mp4"/>
+      <pwccode url="https://github.com/sahaisaumya/informal_fallacies" additional="false">sahaisaumya/informal_fallacies</pwccode>
     </paper>
     <paper id="54">
       <title><fixed-case>S</fixed-case>oc<fixed-case>A</fixed-case>o<fixed-case>G</fixed-case>: Incremental Graph Parsing for Social Relation Inference in Dialogues</title>
@@ -855,8 +856,8 @@
       <url hash="b72115f3">2021.acl-long.54</url>
       <doi>10.18653/v1/2021.acl-long.54</doi>
       <bibkey>qiu-etal-2021-socaog</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/dialogre">DialogRE</pwcdataset>
       <video href="2021.acl-long.54.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/dialogre">DialogRE</pwcdataset>
     </paper>
     <paper id="55">
       <title><fixed-case>T</fixed-case>icket<fixed-case>T</fixed-case>alk: Toward human-level performance with end-to-end, transaction-based dialog systems</title>
@@ -869,10 +870,10 @@
       <url hash="0cd52be3">2021.acl-long.55</url>
       <doi>10.18653/v1/2021.acl-long.55</doi>
       <bibkey>byrne-etal-2021-tickettalk</bibkey>
+      <video href="2021.acl-long.55.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/tickettalk">TicketTalk</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/taskmaster-2">Taskmaster-2</pwcdataset>
-      <video href="2021.acl-long.55.mp4"/>
     </paper>
     <paper id="56">
       <title>Improving Dialog Systems for Negotiation with Personality Modeling</title>
@@ -884,8 +885,8 @@
       <url hash="d7a905f0">2021.acl-long.56</url>
       <doi>10.18653/v1/2021.acl-long.56</doi>
       <bibkey>yang-etal-2021-improving</bibkey>
-      <pwccode url="https://github.com/princeton-nlp/NegotiationToM" additional="false">princeton-nlp/NegotiationToM</pwccode>
       <video href="2021.acl-long.56.mp4"/>
+      <pwccode url="https://github.com/princeton-nlp/NegotiationToM" additional="false">princeton-nlp/NegotiationToM</pwccode>
     </paper>
     <paper id="57">
       <title>Learning from Perturbations: Diverse and Informative Dialogue Generation with Inverse Adversarial Training</title>
@@ -897,9 +898,9 @@
       <url hash="f64123b6">2021.acl-long.57</url>
       <doi>10.18653/v1/2021.acl-long.57</doi>
       <bibkey>zhou-etal-2021-learning</bibkey>
+      <video href="2021.acl-long.57.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
-      <video href="2021.acl-long.57.mp4"/>
     </paper>
     <paper id="58">
       <title>Increasing Faithfulness in Knowledge-Grounded Dialogue with Controllable Features</title>
@@ -912,8 +913,8 @@
       <url hash="48f596d7">2021.acl-long.58</url>
       <doi>10.18653/v1/2021.acl-long.58</doi>
       <bibkey>rashkin-etal-2021-increasing</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
       <video href="2021.acl-long.58.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
     </paper>
     <paper id="59">
       <title><fixed-case>C</fixed-case>itation<fixed-case>IE</fixed-case>: Leveraging the Citation Graph for Scientific Information Extraction</title>
@@ -925,10 +926,10 @@
       <url hash="f18c417d">2021.acl-long.59</url>
       <doi>10.18653/v1/2021.acl-long.59</doi>
       <bibkey>viswanathan-etal-2021-citationie</bibkey>
+      <video href="2021.acl-long.59.mp4"/>
       <pwccode url="https://github.com/viswavi/ScigraphIE" additional="false">viswavi/ScigraphIE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/s2orc">S2ORC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scirex">SciREX</pwcdataset>
-      <video href="2021.acl-long.59.mp4"/>
     </paper>
     <paper id="60">
       <title>From Discourse to Narrative: Knowledge Projection for Event Relation Extraction</title>
@@ -945,10 +946,10 @@
       <url hash="5e4ab92f">2021.acl-long.60</url>
       <doi>10.18653/v1/2021.acl-long.60</doi>
       <bibkey>tang-etal-2021-discourse</bibkey>
+      <video href="2021.acl-long.60.mp4"/>
       <pwccode url="https://github.com/TangJiaLong/Knowledge-Projection-for-ERE" additional="false">TangJiaLong/Knowledge-Projection-for-ERE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
-      <video href="2021.acl-long.60.mp4"/>
     </paper>
     <paper id="61">
       <title><fixed-case>A</fixed-case>dv<fixed-case>P</fixed-case>icker: <fixed-case>E</fixed-case>ffectively <fixed-case>L</fixed-case>everaging <fixed-case>U</fixed-case>nlabeled <fixed-case>D</fixed-case>ata via <fixed-case>A</fixed-case>dversarial <fixed-case>D</fixed-case>iscriminator for <fixed-case>C</fixed-case>ross-<fixed-case>L</fixed-case>ingual <fixed-case>NER</fixed-case></title>
@@ -962,10 +963,10 @@
       <url hash="71ece7d0">2021.acl-long.61</url>
       <doi>10.18653/v1/2021.acl-long.61</doi>
       <bibkey>chen-etal-2021-advpicker</bibkey>
+      <video href="2021.acl-long.61.mp4"/>
       <pwccode url="https://github.com/microsoft/vert-papers/tree/master/papers/AdvPicker" additional="false">microsoft/vert-papers</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2002">CoNLL 2002</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
-      <video href="2021.acl-long.61.mp4"/>
     </paper>
     <paper id="62">
       <title>Compare to The Knowledge: Graph Neural Fake News Detection with External Knowledge</title>
@@ -982,8 +983,8 @@
       <url hash="144fff95">2021.acl-long.62</url>
       <doi>10.18653/v1/2021.acl-long.62</doi>
       <bibkey>hu-etal-2021-compare</bibkey>
-      <pwccode url="https://github.com/ytc272098215/fakenewsdetection" additional="false">ytc272098215/fakenewsdetection</pwccode>
       <video href="2021.acl-long.62.mp4"/>
+      <pwccode url="https://github.com/ytc272098215/fakenewsdetection" additional="false">ytc272098215/fakenewsdetection</pwccode>
     </paper>
     <paper id="63">
       <title>Discontinuous Named Entity Recognition as Maximal Clique Discovery</title>
@@ -998,8 +999,8 @@
       <url hash="3a68f07b">2021.acl-long.63</url>
       <doi>10.18653/v1/2021.acl-long.63</doi>
       <bibkey>wang-etal-2021-discontinuous</bibkey>
-      <pwccode url="https://github.com/131250208/infextraction" additional="false">131250208/infextraction</pwccode>
       <video href="2021.acl-long.63.mp4"/>
+      <pwccode url="https://github.com/131250208/infextraction" additional="false">131250208/infextraction</pwccode>
     </paper>
     <paper id="64">
       <title><fixed-case>LNN</fixed-case>-<fixed-case>EL</fixed-case>: A Neuro-Symbolic Approach to Short-text Entity Linking</title>
@@ -1016,8 +1017,8 @@
       <url hash="ed3dd619">2021.acl-long.64</url>
       <doi>10.18653/v1/2021.acl-long.64</doi>
       <bibkey>jiang-etal-2021-lnn</bibkey>
-      <pwccode url="https://github.com/IBM/LNN" additional="false">IBM/LNN</pwccode>
       <video href="2021.acl-long.64.mp4"/>
+      <pwccode url="https://github.com/IBM/LNN" additional="false">IBM/LNN</pwccode>
     </paper>
     <paper id="65">
       <title>Do Context-Aware Translation Models Pay the Right Attention?</title>
@@ -1032,8 +1033,8 @@
       <url hash="173c6b8e">2021.acl-long.65</url>
       <doi>10.18653/v1/2021.acl-long.65</doi>
       <bibkey>yin-etal-2021-context</bibkey>
-      <pwccode url="https://github.com/neulab/contextual-mt" additional="false">neulab/contextual-mt</pwccode>
       <video href="2021.acl-long.65.mp4"/>
+      <pwccode url="https://github.com/neulab/contextual-mt" additional="false">neulab/contextual-mt</pwccode>
     </paper>
     <paper id="66">
       <title>Adapting High-resource <fixed-case>NMT</fixed-case> Models to Translate Low-resource Related Languages without Parallel Data</title>
@@ -1051,10 +1052,10 @@
       <url hash="3c7907c0">2021.acl-long.66</url>
       <doi>10.18653/v1/2021.acl-long.66</doi>
       <bibkey>ko-etal-2021-adapting</bibkey>
+      <video href="2021.acl-long.66.mp4"/>
       <pwccode url="https://github.com/wjko2/NMT-Adapt" additional="false">wjko2/NMT-Adapt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cc100">CC100</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flores">FLoRes</pwcdataset>
-      <video href="2021.acl-long.66.mp4"/>
     </paper>
     <paper id="67">
       <title>Bilingual Lexicon Induction via Unsupervised Bitext Construction and Word Alignment</title>
@@ -1067,8 +1068,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="8b0bf445">2021.acl-long.67.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.67</doi>
       <bibkey>shi-etal-2021-bilingual</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wikimatrix">WikiMatrix</pwcdataset>
       <video href="2021.acl-long.67.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikimatrix">WikiMatrix</pwcdataset>
     </paper>
     <paper id="68">
       <title>Multilingual Speech Translation from Efficient Finetuning of Pretrained Models</title>
@@ -1086,8 +1087,8 @@
       <url hash="1d7c0fbd">2021.acl-long.68</url>
       <doi>10.18653/v1/2021.acl-long.68</doi>
       <bibkey>li-etal-2021-multilingual</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/europarl-st">Europarl-ST</pwcdataset>
       <video href="2021.acl-long.68.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/europarl-st">Europarl-ST</pwcdataset>
     </paper>
     <paper id="69">
       <title>Learning Faithful Representations of Causal Graphs</title>
@@ -1109,9 +1110,10 @@
       <url hash="2d45ff63">2021.acl-long.70</url>
       <doi>10.18653/v1/2021.acl-long.70</doi>
       <bibkey>oconnor-andreas-2021-context</bibkey>
+      <video href="2021.acl-long.70.mp4"/>
       <pwccode url="https://github.com/lingo-mit/context-ablations" additional="false">lingo-mit/context-ablations</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.acl-long.70.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="71">
       <title>Integrated Directional Gradients: Feature Interaction Attribution for Neural <fixed-case>NLP</fixed-case> Models</title>
@@ -1123,10 +1125,10 @@
       <url hash="8643e824">2021.acl-long.71</url>
       <doi>10.18653/v1/2021.acl-long.71</doi>
       <bibkey>sikdar-etal-2021-integrated</bibkey>
+      <video href="2021.acl-long.71.mp4"/>
       <pwccode url="https://github.com/parantapa/integrated-directional-gradients" additional="false">parantapa/integrated-directional-gradients</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.71.mp4"/>
     </paper>
     <paper id="72">
       <title><fixed-case>D</fixed-case>e<fixed-case>CLUTR</fixed-case>: Deep Contrastive Learning for Unsupervised Textual Representations</title>
@@ -1139,6 +1141,7 @@
       <url hash="0b3cbb4c">2021.acl-long.72</url>
       <doi>10.18653/v1/2021.acl-long.72</doi>
       <bibkey>giorgi-etal-2021-declutr</bibkey>
+      <video href="2021.acl-long.72.mp4"/>
       <pwccode url="https://github.com/JohnGiorgi/DeCLUTR" additional="true">JohnGiorgi/DeCLUTR</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
@@ -1147,7 +1150,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.acl-long.72.mp4"/>
     </paper>
     <paper id="73">
       <title><fixed-case>XLPT</fixed-case>-<fixed-case>AMR</fixed-case>: Cross-Lingual Pre-Training via Multi-Task Learning for Zero-Shot <fixed-case>AMR</fixed-case> Parsing and Text Generation</title>
@@ -1172,10 +1174,10 @@
       <url hash="f4d1696f">2021.acl-long.74</url>
       <doi>10.18653/v1/2021.acl-long.74</doi>
       <bibkey>herzig-berant-2021-span</bibkey>
+      <video href="2021.acl-long.74.mp4"/>
       <pwccode url="https://github.com/jonathanherzig/span-based-sp" additional="false">jonathanherzig/span-based-sp</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clevr">CLEVR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
-      <video href="2021.acl-long.74.mp4"/>
     </paper>
     <paper id="75">
       <title>Compositional Generalization and Natural Language Variation: Can a Semantic Parsing Approach Handle Both?</title>
@@ -1188,10 +1190,10 @@
       <url hash="12cb5824">2021.acl-long.75</url>
       <doi>10.18653/v1/2021.acl-long.75</doi>
       <bibkey>shaw-etal-2021-compositional</bibkey>
+      <video href="2021.acl-long.75.mp4"/>
       <pwccode url="https://github.com/google-research/language/tree/master/language/nqg" additional="false">google-research/language</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
-      <video href="2021.acl-long.75.mp4"/>
     </paper>
     <paper id="76">
       <title>A Targeted Assessment of Incremental Processing in Neural Language Models and Humans</title>
@@ -1203,8 +1205,8 @@
       <url hash="1adfcc66">2021.acl-long.76</url>
       <doi>10.18653/v1/2021.acl-long.76</doi>
       <bibkey>wilcox-etal-2021-targeted</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/billion-word-benchmark">Billion Word Benchmark</pwcdataset>
       <video href="2021.acl-long.76.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/billion-word-benchmark">Billion Word Benchmark</pwcdataset>
     </paper>
     <paper id="77">
       <title>The Possible, the Plausible, and the Desirable: Event-Based Modality Detection for Language Processing</title>
@@ -1218,8 +1220,8 @@
       <url hash="fba8beda">2021.acl-long.77</url>
       <doi>10.18653/v1/2021.acl-long.77</doi>
       <bibkey>pyatkin-etal-2021-possible</bibkey>
-      <pwccode url="https://github.com/OnlpLab/Modality-Corpus" additional="true">OnlpLab/Modality-Corpus</pwccode>
       <video href="2021.acl-long.77.mp4"/>
+      <pwccode url="https://github.com/OnlpLab/Modality-Corpus" additional="true">OnlpLab/Modality-Corpus</pwccode>
     </paper>
     <paper id="78">
       <title>To <fixed-case>POS</fixed-case> Tag or Not to <fixed-case>POS</fixed-case> Tag: The Impact of <fixed-case>POS</fixed-case> Tags on Morphological Learning in Low-Resource Settings</title>
@@ -1264,12 +1266,12 @@
       <url hash="be31e7e8">2021.acl-long.80</url>
       <doi>10.18653/v1/2021.acl-long.80</doi>
       <bibkey>wang-etal-2021-voxpopuli</bibkey>
+      <video href="2021.acl-long.80.mp4"/>
       <pwccode url="https://github.com/facebookresearch/voxpopuli" additional="false">facebookresearch/voxpopuli</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/voxpopuli">VoxPopuli</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/common-voice">Common Voice</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/europarl-st">Europarl-ST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
-      <video href="2021.acl-long.80.mp4"/>
     </paper>
     <paper id="81">
       <title>Stereotyping <fixed-case>N</fixed-case>orwegian Salmon: An Inventory of Pitfalls in Fairness Benchmark Datasets</title>
@@ -1283,10 +1285,10 @@
       <url hash="5a9672f9">2021.acl-long.81</url>
       <doi>10.18653/v1/2021.acl-long.81</doi>
       <bibkey>blodgett-etal-2021-stereotyping</bibkey>
+      <video href="2021.acl-long.81.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/crows-pairs">CrowS-Pairs</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/stereoset">StereoSet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winobias">WinoBias</pwcdataset>
-      <video href="2021.acl-long.81.mp4"/>
     </paper>
     <paper id="82">
       <title>Robust Knowledge Graph Completion with Stacked Convolutions and a Student Re-Ranking Network</title>
@@ -1300,10 +1302,10 @@
       <url hash="e1522f5e">2021.acl-long.82</url>
       <doi>10.18653/v1/2021.acl-long.82</doi>
       <bibkey>lovelace-etal-2021-robust</bibkey>
+      <video href="2021.acl-long.82.mp4"/>
       <pwccode url="https://github.com/justinlovelace/robust-kg-completion" additional="false">justinlovelace/robust-kg-completion</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/umls">UMLS</pwcdataset>
-      <video href="2021.acl-long.82.mp4"/>
     </paper>
     <paper id="83">
       <title>A <fixed-case>DQN</fixed-case>-based Approach to Finding Precise Evidences for Fact Verification</title>
@@ -1318,9 +1320,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="30248ef7">2021.acl-long.83.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.83</doi>
       <bibkey>wan-etal-2021-dqn</bibkey>
+      <video href="2021.acl-long.83.mp4"/>
       <pwccode url="https://github.com/sysulic/dqn-fv" additional="false">sysulic/dqn-fv</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
-      <video href="2021.acl-long.83.mp4"/>
     </paper>
     <paper id="84">
       <title>The Art of Abstention: Selective Prediction and Error Regularization for Natural Language Processing</title>
@@ -1333,12 +1335,12 @@
       <url hash="f786865b">2021.acl-long.84</url>
       <doi>10.18653/v1/2021.acl-long.84</doi>
       <bibkey>xin-etal-2021-art</bibkey>
+      <video href="2021.acl-long.84.mp4"/>
       <pwccode url="https://github.com/castorini/transformers-selective" additional="false">castorini/transformers-selective</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.84.mp4"/>
     </paper>
     <paper id="85">
       <title>Unsupervised Out-of-Domain Detection via Pre-trained Transformers</title>
@@ -1352,10 +1354,10 @@
       <url hash="ca314d65">2021.acl-long.85</url>
       <doi>10.18653/v1/2021.acl-long.85</doi>
       <bibkey>xu-etal-2021-unsupervised</bibkey>
+      <video href="2021.acl-long.85.mp4"/>
       <pwccode url="https://github.com/rivercold/BERT-unsupervised-OOD" additional="false">rivercold/BERT-unsupervised-OOD</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.85.mp4"/>
     </paper>
     <paper id="86">
       <title><fixed-case>MATE</fixed-case>-<fixed-case>KD</fixed-case>: Masked Adversarial <fixed-case>TE</fixed-case>xt, a Companion to Knowledge Distillation</title>
@@ -1367,10 +1369,10 @@
       <url hash="154d9421">2021.acl-long.86</url>
       <doi>10.18653/v1/2021.acl-long.86</doi>
       <bibkey>rashid-etal-2021-mate</bibkey>
+      <video href="2021.acl-long.86.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws">PAWS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.86.mp4"/>
     </paper>
     <paper id="87">
       <title>Selecting Informative Contexts Improves Language Model Fine-tuning</title>
@@ -1383,9 +1385,10 @@
       <url hash="45a749fb">2021.acl-long.87</url>
       <doi>10.18653/v1/2021.acl-long.87</doi>
       <bibkey>antonello-etal-2021-selecting</bibkey>
+      <video href="2021.acl-long.87.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.acl-long.87.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="88">
       <title>Explainable Prediction of Text Complexity: The Missing Preliminaries for Text Simplification</title>
@@ -1398,9 +1401,9 @@
       <url hash="9820d084">2021.acl-long.88</url>
       <doi>10.18653/v1/2021.acl-long.88</doi>
       <bibkey>garbacea-etal-2021-explainable</bibkey>
+      <video href="2021.acl-long.88.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/newsela">Newsela</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikilarge">WikiLarge</pwcdataset>
-      <video href="2021.acl-long.88.mp4"/>
     </paper>
     <paper id="89">
       <title>Multi-Task Retrieval for Knowledge-Intensive Tasks</title>
@@ -1416,11 +1419,11 @@
       <url hash="71846e23">2021.acl-long.89</url>
       <doi>10.18653/v1/2021.acl-long.89</doi>
       <bibkey>maillard-etal-2021-multi</bibkey>
+      <video href="2021.acl-long.89.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.acl-long.89.mp4"/>
     </paper>
     <paper id="90">
       <title>When Do You Need Billions of Words of Pretraining Data?</title>
@@ -1433,13 +1436,13 @@
       <url hash="a10434a4">2021.acl-long.90</url>
       <doi>10.18653/v1/2021.acl-long.90</doi>
       <bibkey>zhang-etal-2021-need</bibkey>
+      <video href="2021.acl-long.90.mp4"/>
       <pwccode url="https://github.com/nyu-mll/pretraining-learning-curves" additional="false">nyu-mll/pretraining-learning-curves</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/blimp">BLiMP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
-      <video href="2021.acl-long.90.mp4"/>
     </paper>
     <paper id="91">
       <title>Analyzing the Source and Target Contributions to Predictions in Neural Machine Translation</title>
@@ -1451,8 +1454,8 @@
       <url hash="b360eb57">2021.acl-long.91</url>
       <doi>10.18653/v1/2021.acl-long.91</doi>
       <bibkey>voita-etal-2021-analyzing</bibkey>
-      <pwccode url="https://github.com/lena-voita/the-story-of-heads" additional="false">lena-voita/the-story-of-heads</pwccode>
       <video href="2021.acl-long.91.mp4"/>
+      <pwccode url="https://github.com/lena-voita/the-story-of-heads" additional="false">lena-voita/the-story-of-heads</pwccode>
     </paper>
     <paper id="92">
       <title>Comparing Test Sets with Item Response Theory</title>
@@ -1470,6 +1473,7 @@
       <url hash="8d629013">2021.acl-long.92</url>
       <doi>10.18653/v1/2021.acl-long.92</doi>
       <bibkey>vania-etal-2021-comparing</bibkey>
+      <video href="2021.acl-long.92.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/commonsenseqa">CommonsenseQA</pwcdataset>
@@ -1489,7 +1493,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winogrande">WinoGrande</pwcdataset>
-      <video href="2021.acl-long.92.mp4"/>
     </paper>
     <paper id="93">
       <title>Uncovering Constraint-Based Behavior in Neural Models via Targeted Fine-Tuning</title>
@@ -1500,8 +1503,8 @@
       <url hash="ed3497d7">2021.acl-long.93</url>
       <doi>10.18653/v1/2021.acl-long.93</doi>
       <bibkey>davis-van-schijndel-2021-uncovering</bibkey>
-      <pwccode url="https://github.com/forrestdavis/ImplicitCausality" additional="false">forrestdavis/ImplicitCausality</pwccode>
       <video href="2021.acl-long.93.mp4"/>
+      <pwccode url="https://github.com/forrestdavis/ImplicitCausality" additional="false">forrestdavis/ImplicitCausality</pwccode>
     </paper>
     <paper id="94">
       <title>More Identifiable yet Equally Performant Transformers for Text Classification</title>
@@ -1514,12 +1517,12 @@
       <url hash="7d1720ba">2021.acl-long.94</url>
       <doi>10.18653/v1/2021.acl-long.94</doi>
       <bibkey>bhardwaj-etal-2021-identifiable</bibkey>
+      <video href="2021.acl-long.94.mp4"/>
       <pwccode url="https://github.com/declare-lab/identifiable-transformers" additional="false">declare-lab/identifiable-transformers</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.94.mp4"/>
     </paper>
     <paper id="95">
       <title><fixed-case>A</fixed-case>ug<fixed-case>NLG</fixed-case>: Few-shot Natural Language Generation using Self-trained Data Augmentation</title>
@@ -1532,8 +1535,8 @@
       <url hash="f0499e65">2021.acl-long.95</url>
       <doi>10.18653/v1/2021.acl-long.95</doi>
       <bibkey>xu-etal-2021-augnlg</bibkey>
-      <pwccode url="https://github.com/XinnuoXu/AugNLG" additional="false">XinnuoXu/AugNLG</pwccode>
       <video href="2021.acl-long.95.mp4"/>
+      <pwccode url="https://github.com/XinnuoXu/AugNLG" additional="false">XinnuoXu/AugNLG</pwccode>
     </paper>
     <paper id="96">
       <title>Can vectors read minds better than experts? Comparing data augmentation strategies for the automated scoring of children’s mindreading ability</title>
@@ -1561,9 +1564,9 @@
       <url hash="e60b443b">2021.acl-long.97</url>
       <doi>10.18653/v1/2021.acl-long.97</doi>
       <bibkey>zhang-etal-2021-dataset</bibkey>
+      <video href="2021.acl-long.97.mp4"/>
       <pwccode url="https://github.com/zhangmozhi/mrs" additional="false">zhangmozhi/mrs</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mrs">MRS</pwcdataset>
-      <video href="2021.acl-long.97.mp4"/>
     </paper>
     <paper id="98">
       <title>What Ingredients Make for an Effective Crowdsourcing Protocol for Difficult <fixed-case>NLU</fixed-case> Data Collection Tasks?</title>
@@ -1579,10 +1582,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="1261b301">2021.acl-long.98.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.98</doi>
       <bibkey>nangia-etal-2021-ingredients</bibkey>
+      <video href="2021.acl-long.98.mp4"/>
       <pwccode url="https://github.com/nyu-mll/crowdsourcing-protocol-comparison" additional="false">nyu-mll/crowdsourcing-protocol-comparison</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cosmosqa">CosmosQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
-      <video href="2021.acl-long.98.mp4"/>
     </paper>
     <paper id="99">
       <title>Align Voting Behavior with Public Statements for Legislator Representation Learning</title>
@@ -1624,9 +1627,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="e6ca9940">2021.acl-long.101.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.101</doi>
       <bibkey>liu-etal-2021-improving-zero</bibkey>
+      <video href="2021.acl-long.101.mp4"/>
       <pwccode url="https://github.com/nlp-dke/NMTGMinor/tree/master/recipes/zero-shot" additional="false">nlp-dke/NMTGMinor</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/pmindia">PMIndia</pwcdataset>
-      <video href="2021.acl-long.101.mp4"/>
     </paper>
     <paper id="102">
       <title>Common Sense Beyond <fixed-case>E</fixed-case>nglish: Evaluating and Improving Multilingual Language Models for Commonsense Reasoning</title>
@@ -1639,6 +1642,7 @@
       <url hash="5ae6024c">2021.acl-long.102</url>
       <doi>10.18653/v1/2021.acl-long.102</doi>
       <bibkey>lin-etal-2021-common</bibkey>
+      <video href="2021.acl-long.102.mp4"/>
       <pwccode url="https://github.com/INK-USC/XCSR" additional="false">INK-USC/XCSR</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/x-csqa">X-CSQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cc100">CC100</pwcdataset>
@@ -1647,7 +1651,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/swag">SWAG</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.acl-long.102.mp4"/>
     </paper>
     <paper id="103">
       <title>Attention Calibration for Transformer in Neural Machine Translation</title>
@@ -1689,8 +1692,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="405e351d">2021.acl-long.105.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.105</doi>
       <bibkey>khemchandani-etal-2021-exploiting</bibkey>
-      <pwccode url="https://github.com/yashkhem1/RelateLM" additional="false">yashkhem1/RelateLM</pwccode>
       <video href="2021.acl-long.105.mp4"/>
+      <pwccode url="https://github.com/yashkhem1/RelateLM" additional="false">yashkhem1/RelateLM</pwccode>
     </paper>
     <paper id="106">
       <title>On Finding the K-best Non-projective Dependency Trees</title>
@@ -1702,8 +1705,8 @@
       <url hash="b2d321c9">2021.acl-long.106</url>
       <doi>10.18653/v1/2021.acl-long.106</doi>
       <bibkey>zmigrod-etal-2021-finding</bibkey>
-      <pwccode url="https://github.com/rycolab/spanningtrees" additional="false">rycolab/spanningtrees</pwccode>
       <video href="2021.acl-long.106.mp4"/>
+      <pwccode url="https://github.com/rycolab/spanningtrees" additional="false">rycolab/spanningtrees</pwccode>
     </paper>
     <paper id="107">
       <title>Towards Argument Mining for Social Good: A Survey</title>
@@ -1727,8 +1730,8 @@
       <url hash="63bc0965">2021.acl-long.108</url>
       <doi>10.18653/v1/2021.acl-long.108</doi>
       <bibkey>fontana-de-vargas-moffatt-2021-automated</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/vist">VIST</pwcdataset>
       <video href="2021.acl-long.108.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/vist">VIST</pwcdataset>
     </paper>
     <paper id="109">
       <title><fixed-case>CLIP</fixed-case>: A Dataset for Extracting Action Items for Physicians from Hospital Discharge Notes</title>
@@ -1746,10 +1749,10 @@
       <url hash="e228f064">2021.acl-long.109</url>
       <doi>10.18653/v1/2021.acl-long.109</doi>
       <bibkey>mullenbach-etal-2021-clip</bibkey>
+      <video href="2021.acl-long.109.mp4"/>
       <pwccode url="https://github.com/asappresearch/clip" additional="false">asappresearch/clip</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clip">CLIP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mimic-iii">MIMIC-III</pwcdataset>
-      <video href="2021.acl-long.109.mp4"/>
     </paper>
     <paper id="110">
       <title>Assessing Emoji Use in Modern Text Processing Tools</title>
@@ -1786,11 +1789,11 @@
       <url hash="79b041c0">2021.acl-long.112</url>
       <doi>10.18653/v1/2021.acl-long.112</doi>
       <bibkey>hosking-lapata-2021-factorising</bibkey>
+      <video href="2021.acl-long.112.mp4"/>
       <pwccode url="https://github.com/tomhosking/separator" additional="false">tomhosking/separator</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paralex">Paralex</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quora-question-pairs">Quora Question Pairs</pwcdataset>
-      <video href="2021.acl-long.112.mp4"/>
     </paper>
     <paper id="113">
       <title><fixed-case>A</fixed-case>gg<fixed-case>G</fixed-case>en: Ordering and Aggregating while Generating</title>
@@ -1803,9 +1806,9 @@
       <url hash="f14d5d37">2021.acl-long.113</url>
       <doi>10.18653/v1/2021.acl-long.113</doi>
       <bibkey>xu-etal-2021-agggen</bibkey>
+      <video href="2021.acl-long.113.mp4"/>
       <pwccode url="https://github.com/XinnuoXu/AggGen" additional="false">XinnuoXu/AggGen</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.acl-long.113.mp4"/>
     </paper>
     <paper id="114">
       <title>Reflective Decoding: Beyond Unidirectional Generation with Off-the-Shelf Language Models</title>
@@ -1834,9 +1837,9 @@
       <url hash="92301d86">2021.acl-long.115</url>
       <doi>10.18653/v1/2021.acl-long.115</doi>
       <bibkey>suadaa-etal-2021-towards</bibkey>
+      <video href="2021.acl-long.115.mp4"/>
       <pwccode url="https://github.com/titech-nlp/numeric-nlg" additional="false">titech-nlp/numeric-nlg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/rotowire">RotoWire</pwcdataset>
-      <video href="2021.acl-long.115.mp4"/>
     </paper>
     <paper id="116">
       <title><fixed-case>BACO</fixed-case>: A Background Knowledge- and Content-Based Framework for Citing Sentence Generation</title>
@@ -1867,8 +1870,8 @@
       <url hash="9b8b8ffa">2021.acl-long.117</url>
       <doi>10.18653/v1/2021.acl-long.117</doi>
       <bibkey>feng-etal-2021-language</bibkey>
-      <pwccode url="https://github.com/xcfcode/PLM_annotator" additional="false">xcfcode/PLM_annotator</pwccode>
       <video href="2021.acl-long.117.mp4"/>
+      <pwccode url="https://github.com/xcfcode/PLM_annotator" additional="false">xcfcode/PLM_annotator</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
     </paper>
     <paper id="118">
@@ -1880,10 +1883,10 @@
       <url hash="650fb73f">2021.acl-long.118</url>
       <doi>10.18653/v1/2021.acl-long.118</doi>
       <bibkey>asai-choi-2021-challenges</bibkey>
+      <video href="2021.acl-long.118.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coqa">CoQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
-      <video href="2021.acl-long.118.mp4"/>
     </paper>
     <paper id="119">
       <title>A Gradually Soft Multi-Task and Data-Augmented Approach to Medical Question Understanding</title>
@@ -1899,10 +1902,10 @@
       <url hash="df3b64a6">2021.acl-long.119</url>
       <doi>10.18653/v1/2021.acl-long.119</doi>
       <bibkey>mrini-etal-2021-gradually</bibkey>
+      <video href="2021.acl-long.119.mp4"/>
       <pwccode url="https://github.com/khalilmrini/medical-question-understanding" additional="false">khalilmrini/medical-question-understanding</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meqsum">MeQSum</pwcdataset>
-      <video href="2021.acl-long.119.mp4"/>
     </paper>
     <paper id="120">
       <title>Leveraging Type Descriptions for Zero-shot Named Entity Recognition and Classification</title>
@@ -1914,8 +1917,8 @@
       <url hash="b8af64ea">2021.acl-long.120</url>
       <doi>10.18653/v1/2021.acl-long.120</doi>
       <bibkey>aly-etal-2021-leveraging</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/medmentions">MedMentions</pwcdataset>
       <video href="2021.acl-long.120.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/medmentions">MedMentions</pwcdataset>
     </paper>
     <paper id="121">
       <title><fixed-case>MECT</fixed-case>: <fixed-case>M</fixed-case>ulti-Metadata Embedding based Cross-Transformer for <fixed-case>C</fixed-case>hinese Named Entity Recognition</title>
@@ -1927,8 +1930,8 @@
       <url hash="dc3a2cf5">2021.acl-long.121</url>
       <doi>10.18653/v1/2021.acl-long.121</doi>
       <bibkey>wu-etal-2021-mect</bibkey>
-      <pwccode url="https://github.com/CoderMusou/MECT4CNER" additional="false">CoderMusou/MECT4CNER</pwccode>
       <video href="2021.acl-long.121.mp4"/>
+      <pwccode url="https://github.com/CoderMusou/MECT4CNER" additional="false">CoderMusou/MECT4CNER</pwccode>
     </paper>
     <paper id="122">
       <title>Factuality Assessment as Modal Dependency Parsing</title>
@@ -1942,8 +1945,8 @@
       <url hash="94d087d1">2021.acl-long.122</url>
       <doi>10.18653/v1/2021.acl-long.122</doi>
       <bibkey>yao-etal-2021-factuality</bibkey>
-      <pwccode url="https://github.com/jryao/modal_dependency" additional="false">jryao/modal_dependency</pwccode>
       <video href="2021.acl-long.122.mp4"/>
+      <pwccode url="https://github.com/jryao/modal_dependency" additional="false">jryao/modal_dependency</pwccode>
     </paper>
     <paper id="123">
       <title>Directed Acyclic Graph Network for Conversational Emotion Recognition</title>
@@ -1956,12 +1959,12 @@
       <url hash="7762d3e5">2021.acl-long.123</url>
       <doi>10.18653/v1/2021.acl-long.123</doi>
       <bibkey>shen-etal-2021-directed</bibkey>
+      <video href="2021.acl-long.123.mp4"/>
       <pwccode url="https://github.com/shenwzh3/DAG-ERC" additional="false">shenwzh3/DAG-ERC</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/emorynlp">EmoryNLP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
-      <video href="2021.acl-long.123.mp4"/>
     </paper>
     <paper id="124">
       <title>Improving Formality Style Transfer with Context-Aware Rule Injection</title>
@@ -1972,8 +1975,8 @@
       <url hash="5ddea311">2021.acl-long.124</url>
       <doi>10.18653/v1/2021.acl-long.124</doi>
       <bibkey>yao-yu-2021-improving</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
       <video href="2021.acl-long.124.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
     </paper>
     <paper id="125">
       <title>Topic-Driven and Knowledge-Aware Transformer for Dialogue Emotion Detection</title>
@@ -1987,13 +1990,13 @@
       <url hash="f97a3caa">2021.acl-long.125</url>
       <doi>10.18653/v1/2021.acl-long.125</doi>
       <bibkey>zhu-etal-2021-topic</bibkey>
+      <video href="2021.acl-long.125.mp4"/>
       <pwccode url="https://github.com/something678/TodKat" additional="false">something678/TodKat</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/emorynlp">EmoryNLP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
-      <video href="2021.acl-long.125.mp4"/>
     </paper>
     <paper id="126">
       <title>Syntopical Graphs for Computational Argumentation Tasks</title>
@@ -2039,9 +2042,9 @@
       <url hash="e90c3419">2021.acl-long.128</url>
       <doi>10.18653/v1/2021.acl-long.128</doi>
       <bibkey>si-etal-2021-topic</bibkey>
+      <video href="2021.acl-long.128.mp4"/>
       <pwccode url="https://github.com/jasenchn/TARSA" additional="false">jasenchn/TARSA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
-      <video href="2021.acl-long.128.mp4"/>
     </paper>
     <paper id="129">
       <title>Changes in <fixed-case>E</fixed-case>uropean Solidarity Before and During <fixed-case>COVID</fixed-case>-19: Evidence from a Large Crowd- and Expert-Annotated <fixed-case>T</fixed-case>witter Dataset</title>
@@ -2054,8 +2057,8 @@
       <url hash="04251e57">2021.acl-long.129</url>
       <doi>10.18653/v1/2021.acl-long.129</doi>
       <bibkey>ils-etal-2021-changes</bibkey>
-      <pwccode url="https://github.com/lalashiwoya/socialsolidarityCOVID19" additional="false">lalashiwoya/socialsolidarityCOVID19</pwccode>
       <video href="2021.acl-long.129.mp4"/>
+      <pwccode url="https://github.com/lalashiwoya/socialsolidarityCOVID19" additional="false">lalashiwoya/socialsolidarityCOVID19</pwccode>
     </paper>
     <paper id="130">
       <title>Measuring Conversational Uptake: A Case Study on Student-Teacher Interactions</title>
@@ -2071,8 +2074,8 @@
       <url hash="d6a5f430">2021.acl-long.130</url>
       <doi>10.18653/v1/2021.acl-long.130</doi>
       <bibkey>demszky-etal-2021-measuring</bibkey>
-      <pwccode url="https://github.com/ddemszky/conversational-uptake" additional="false">ddemszky/conversational-uptake</pwccode>
       <video href="2021.acl-long.130.mp4"/>
+      <pwccode url="https://github.com/ddemszky/conversational-uptake" additional="false">ddemszky/conversational-uptake</pwccode>
     </paper>
     <paper id="131">
       <title>A Survey of Code-switching: Linguistic and Social Perspectives for Language Technologies</title>
@@ -2099,8 +2102,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="adcab28a">2021.acl-long.132.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.132</doi>
       <bibkey>vidgen-etal-2021-learning</bibkey>
-      <pwccode url="https://github.com/bvidgen/Dynamically-Generated-Hate-Speech-Dataset" additional="true">bvidgen/Dynamically-Generated-Hate-Speech-Dataset</pwccode>
       <video href="2021.acl-long.132.mp4"/>
+      <pwccode url="https://github.com/bvidgen/Dynamically-Generated-Hate-Speech-Dataset" additional="true">bvidgen/Dynamically-Generated-Hate-Speech-Dataset</pwccode>
     </paper>
     <paper id="133">
       <title><fixed-case>I</fixed-case>nfo<fixed-case>S</fixed-case>urgeon: Cross-Media Fine-grained Information Consistency Checking for Fake News Detection</title>
@@ -2118,9 +2121,9 @@
       <url hash="f4d7d561">2021.acl-long.133</url>
       <doi>10.18653/v1/2021.acl-long.133</doi>
       <bibkey>fung-etal-2021-infosurgeon</bibkey>
+      <video href="2021.acl-long.133.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/neuralnews">NeuralNews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/realnews">RealNews</pwcdataset>
-      <video href="2021.acl-long.133.mp4"/>
     </paper>
     <paper id="134">
       <title><fixed-case>I</fixed-case> like fish, especially dolphins: Addressing Contradictions in Dialogue Modeling</title>
@@ -2135,13 +2138,13 @@
       <attachment type="OptionalSupplementaryMaterial" hash="aa97f771">2021.acl-long.134.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.134</doi>
       <bibkey>nie-etal-2021-like</bibkey>
+      <video href="2021.acl-long.134.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/blended-skill-talk">Blended Skill Talk</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/convai2">ConvAI2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.acl-long.134.mp4"/>
     </paper>
     <paper id="135">
       <title>A Sequence-to-Sequence Approach to Dialogue State Tracking</title>
@@ -2153,13 +2156,13 @@
       <url hash="8729c293">2021.acl-long.135</url>
       <doi>10.18653/v1/2021.acl-long.135</doi>
       <bibkey>feng-etal-2021-sequence</bibkey>
+      <video href="2021.acl-long.135.mp4"/>
       <pwccode url="https://github.com/sweetalyssum/Seq2Seq-DU" additional="false">sweetalyssum/Seq2Seq-DU</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dialogue-state-tracking-challenge">Dialogue State Tracking Challenge</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sgd">SGD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-oz">Wizard-of-Oz</pwcdataset>
-      <video href="2021.acl-long.135.mp4"/>
     </paper>
     <paper id="136">
       <title>Discovering Dialog Structure Graph for Coherent Dialog Generation</title>
@@ -2174,8 +2177,8 @@
       <url hash="e96e084b">2021.acl-long.136</url>
       <doi>10.18653/v1/2021.acl-long.136</doi>
       <bibkey>xu-etal-2021-discovering</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/douban">Douban</pwcdataset>
       <video href="2021.acl-long.136.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/douban">Douban</pwcdataset>
     </paper>
     <paper id="137">
       <title>Dialogue Response Selection with Hierarchical Curriculum Learning</title>
@@ -2193,11 +2196,11 @@
       <url hash="7b9ac232">2021.acl-long.137</url>
       <doi>10.18653/v1/2021.acl-long.137</doi>
       <bibkey>su-etal-2021-dialogue</bibkey>
+      <video href="2021.acl-long.137.mp4"/>
       <pwccode url="https://github.com/yxuansu/HCL" additional="false">yxuansu/HCL</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/douban">Douban</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/e-commerce-1">E-commerce</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rrs">RRS</pwcdataset>
-      <video href="2021.acl-long.137.mp4"/>
     </paper>
     <paper id="138">
       <title>A Joint Model for Dropped Pronoun Recovery and Conversational Discourse Parsing in <fixed-case>C</fixed-case>hinese Conversational Speech</title>
@@ -2214,8 +2217,8 @@
       <url hash="0e8e1654">2021.acl-long.138</url>
       <doi>10.18653/v1/2021.acl-long.138</doi>
       <bibkey>yang-etal-2021-joint</bibkey>
-      <pwccode url="https://github.com/ningningyang/DiscProReco" additional="false">ningningyang/DiscProReco</pwccode>
       <video href="2021.acl-long.138.mp4"/>
+      <pwccode url="https://github.com/ningningyang/DiscProReco" additional="false">ningningyang/DiscProReco</pwccode>
     </paper>
     <paper id="139">
       <title>A Systematic Investigation of <fixed-case>KB</fixed-case>-Text Embedding Alignment at Scale</title>
@@ -2231,8 +2234,8 @@
       <url hash="ef793981">2021.acl-long.139</url>
       <doi>10.18653/v1/2021.acl-long.139</doi>
       <bibkey>pahuja-etal-2021-systematic</bibkey>
-      <pwccode url="https://github.com/dki-lab/joint-kb-text-embedding" additional="false">dki-lab/joint-kb-text-embedding</pwccode>
       <video href="2021.acl-long.139.mp4"/>
+      <pwccode url="https://github.com/dki-lab/joint-kb-text-embedding" additional="false">dki-lab/joint-kb-text-embedding</pwccode>
     </paper>
     <paper id="140">
       <title>Named Entity Recognition with Small Strongly Labeled and Large Weakly Labeled Data</title>
@@ -2246,9 +2249,9 @@
       <url hash="7295a820">2021.acl-long.140</url>
       <doi>10.18653/v1/2021.acl-long.140</doi>
       <bibkey>jiang-etal-2021-named</bibkey>
+      <video href="2021.acl-long.140.mp4"/>
       <pwccode url="https://github.com/amzn/amazon-weak-ner-needle" additional="false">amzn/amazon-weak-ner-needle</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ncbi-disease-1">NCBI Disease</pwcdataset>
-      <video href="2021.acl-long.140.mp4"/>
     </paper>
     <paper id="141">
       <title>Ultra-Fine Entity Typing with Weak Supervision from a Masked Language Model</title>
@@ -2260,9 +2263,9 @@
       <url hash="55433818">2021.acl-long.141</url>
       <doi>10.18653/v1/2021.acl-long.141</doi>
       <bibkey>dai-etal-2021-ultra</bibkey>
+      <video href="2021.acl-long.141.mp4"/>
       <pwccode url="https://github.com/HKUST-KnowComp/MLMET" additional="false">HKUST-KnowComp/MLMET</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ontonotes-5-0">OntoNotes 5.0</pwcdataset>
-      <video href="2021.acl-long.141.mp4"/>
     </paper>
     <paper id="142">
       <title>Improving Named Entity Recognition by External Context Retrieving and Cooperative Learning</title>
@@ -2278,6 +2281,7 @@
       <url hash="11c5c347">2021.acl-long.142</url>
       <doi>10.18653/v1/2021.acl-long.142</doi>
       <bibkey>wang-etal-2021-improving</bibkey>
+      <video href="2021.acl-long.142.mp4"/>
       <pwccode url="https://github.com/Alibaba-NLP/CLNER" additional="false">Alibaba-NLP/CLNER</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bc5cdr">BC5CDR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll">CoNLL++</pwcdataset>
@@ -2285,7 +2289,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/ncbi-disease-1">NCBI Disease</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wnut-2016-ner">WNUT 2016 NER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wnut-2017-emerging-and-rare-entity">WNUT 2017</pwcdataset>
-      <video href="2021.acl-long.142.mp4"/>
     </paper>
     <paper id="143">
       <title>Implicit Representations of Meaning in Neural Language Models</title>
@@ -2297,8 +2300,8 @@
       <url hash="1bbcb049">2021.acl-long.143</url>
       <doi>10.18653/v1/2021.acl-long.143</doi>
       <bibkey>li-etal-2021-implicit</bibkey>
-      <pwccode url="https://github.com/belindal/state-probes" additional="false">belindal/state-probes</pwccode>
       <video href="2021.acl-long.143.mp4"/>
+      <pwccode url="https://github.com/belindal/state-probes" additional="false">belindal/state-probes</pwccode>
     </paper>
     <paper id="144">
       <title>Causal Analysis of Syntactic Agreement Mechanisms in Neural Language Models</title>
@@ -2313,8 +2316,8 @@
       <url hash="7b8513ca">2021.acl-long.144</url>
       <doi>10.18653/v1/2021.acl-long.144</doi>
       <bibkey>finlayson-etal-2021-causal</bibkey>
-      <pwccode url="https://github.com/mattf1n/lm-intervention" additional="false">mattf1n/lm-intervention</pwccode>
       <video href="2021.acl-long.144.mp4"/>
+      <pwccode url="https://github.com/mattf1n/lm-intervention" additional="false">mattf1n/lm-intervention</pwccode>
     </paper>
     <paper id="145">
       <title>Bird’s Eye: Probing for Linguistic Graph Structures with a Simple Information-Theoretic Approach</title>
@@ -2325,10 +2328,10 @@
       <url hash="757557a8">2021.acl-long.145</url>
       <doi>10.18653/v1/2021.acl-long.145</doi>
       <bibkey>hou-sachan-2021-birds</bibkey>
+      <video href="2021.acl-long.145.mp4"/>
       <pwccode url="https://github.com/yifan-h/Graph_Probe-Birds_Eye" additional="false">yifan-h/Graph_Probe-Birds_Eye</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/amr-bank">AMR Bank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.acl-long.145.mp4"/>
     </paper>
     <paper id="146">
       <title>Knowledgeable or Educated Guess? Revisiting Language Models as Knowledge Bases</title>
@@ -2346,10 +2349,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="64a518e9">2021.acl-long.146.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.146</doi>
       <bibkey>cao-etal-2021-knowledgeable</bibkey>
+      <video href="2021.acl-long.146.mp4"/>
       <pwccode url="https://github.com/c-box/LANKA" additional="false">c-box/LANKA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/commonsenseqa">CommonsenseQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
-      <video href="2021.acl-long.146.mp4"/>
     </paper>
     <paper id="147">
       <title>Poisoning Knowledge Graph Embeddings via Relation Inference Patterns</title>
@@ -2362,9 +2365,9 @@
       <url hash="ecf38add">2021.acl-long.147</url>
       <doi>10.18653/v1/2021.acl-long.147</doi>
       <bibkey>bhardwaj-etal-2021-poisoning</bibkey>
+      <video href="2021.acl-long.147.mp4"/>
       <pwccode url="https://github.com/perubhardwaj/inferenceattack" additional="false">perubhardwaj/inferenceattack</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
-      <video href="2021.acl-long.147.mp4"/>
     </paper>
     <paper id="148">
       <title>Bad Seeds: Evaluating Lexical Methods for Bias Measurement</title>
@@ -2376,8 +2379,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="b92a5e9f">2021.acl-long.148.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.148</doi>
       <bibkey>antoniak-mimno-2021-bad</bibkey>
-      <pwccode url="https://github.com/maria-antoniak/bad-seeds" additional="false">maria-antoniak/bad-seeds</pwccode>
       <video href="2021.acl-long.148.mp4"/>
+      <pwccode url="https://github.com/maria-antoniak/bad-seeds" additional="false">maria-antoniak/bad-seeds</pwccode>
     </paper>
     <paper id="149">
       <title>A Survey of Race, Racism, and Anti-Racism in <fixed-case>NLP</fixed-case></title>
@@ -2390,8 +2393,8 @@
       <url hash="d3e2355f">2021.acl-long.149</url>
       <doi>10.18653/v1/2021.acl-long.149</doi>
       <bibkey>field-etal-2021-survey</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/hate-speech">Hate Speech</pwcdataset>
       <video href="2021.acl-long.149.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/hate-speech">Hate Speech</pwcdataset>
     </paper>
     <paper id="150">
       <title>Intrinsic Bias Metrics Do Not Correlate with Application Bias</title>
@@ -2405,8 +2408,8 @@
       <url hash="03328291">2021.acl-long.150</url>
       <doi>10.18653/v1/2021.acl-long.150</doi>
       <bibkey>goldfarb-tarrant-etal-2021-intrinsic</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/winobias">WinoBias</pwcdataset>
       <video href="2021.acl-long.150.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/winobias">WinoBias</pwcdataset>
     </paper>
     <paper id="151">
       <title><fixed-case>R</fixed-case>eddit<fixed-case>B</fixed-case>ias: A Real-World Resource for Bias Evaluation and Debiasing of Conversational Language Models</title>
@@ -2419,9 +2422,9 @@
       <url hash="ae43a843">2021.acl-long.151</url>
       <doi>10.18653/v1/2021.acl-long.151</doi>
       <bibkey>barikeri-etal-2021-redditbias</bibkey>
+      <video href="2021.acl-long.151.mp4"/>
       <pwccode url="https://github.com/umanlp/RedditBias" additional="false">umanlp/RedditBias</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
-      <video href="2021.acl-long.151.mp4"/>
     </paper>
     <paper id="152">
       <title>Contributions of Transformer Attention Heads in Multi- and Cross-lingual Tasks</title>
@@ -2461,9 +2464,9 @@
       <url hash="5c2cbe02">2021.acl-long.154</url>
       <doi>10.18653/v1/2021.acl-long.154</doi>
       <bibkey>bari-etal-2021-uxla</bibkey>
+      <video href="2021.acl-long.154.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.acl-long.154.mp4"/>
     </paper>
     <paper id="155">
       <title>Glancing Transformer for Non-Autoregressive Neural Machine Translation</title>
@@ -2493,10 +2496,10 @@
       <url hash="1a0bfd08">2021.acl-long.156</url>
       <doi>10.18653/v1/2021.acl-long.156</doi>
       <bibkey>ji-etal-2021-hierarchical</bibkey>
+      <video href="2021.acl-long.156.mp4"/>
       <pwccode url="https://github.com/kirkguo/hcn" additional="false">kirkguo/hcn</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/activitynet-captions">ActivityNet Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/youcook2">YouCook2</pwcdataset>
-      <video href="2021.acl-long.156.mp4"/>
     </paper>
     <paper id="157">
       <title>Control Image Captioning Spatially and Temporally</title>
@@ -2511,8 +2514,8 @@
       <url hash="c897abc2">2021.acl-long.157</url>
       <doi>10.18653/v1/2021.acl-long.157</doi>
       <bibkey>yan-etal-2021-control</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/localized-narratives">Localized Narratives</pwcdataset>
       <video href="2021.acl-long.157.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/localized-narratives">Localized Narratives</pwcdataset>
     </paper>
     <paper id="158">
       <title>Edited Media Understanding Frames: Reasoning About the Intent and Implications of Visual Misinformation</title>
@@ -2528,8 +2531,8 @@
       <url hash="c5c4b557">2021.acl-long.158</url>
       <doi>10.18653/v1/2021.acl-long.158</doi>
       <bibkey>da-etal-2021-edited</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <video href="2021.acl-long.158.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
     </paper>
     <paper id="159">
       <title><fixed-case>PIGL</fixed-case>e<fixed-case>T</fixed-case>: Language Grounding Through Neuro-Symbolic Interaction in a 3<fixed-case>D</fixed-case> World</title>
@@ -2546,8 +2549,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="3ac58209">2021.acl-long.159.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.159</doi>
       <bibkey>zellers-etal-2021-piglet</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/vcr">VCR</pwcdataset>
       <video href="2021.acl-long.159.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/vcr">VCR</pwcdataset>
     </paper>
     <paper id="160">
       <title>Modeling Fine-Grained Entity Types with Box Embeddings</title>
@@ -2560,9 +2563,9 @@
       <url hash="ec470f01">2021.acl-long.160</url>
       <doi>10.18653/v1/2021.acl-long.160</doi>
       <bibkey>onoe-etal-2021-modeling</bibkey>
+      <video href="2021.acl-long.160.mp4"/>
       <pwccode url="https://github.com/yasumasaonoe/Box4Types" additional="false">yasumasaonoe/Box4Types</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/figer">FIGER</pwcdataset>
-      <video href="2021.acl-long.160.mp4"/>
     </paper>
     <paper id="161">
       <title><fixed-case>C</fixed-case>hinese<fixed-case>BERT</fixed-case>: <fixed-case>C</fixed-case>hinese Pretraining Enhanced by Glyph and <fixed-case>P</fixed-case>inyin Information</title>
@@ -2579,12 +2582,12 @@
       <url hash="b1281194">2021.acl-long.161</url>
       <doi>10.18653/v1/2021.acl-long.161</doi>
       <bibkey>sun-etal-2021-chinesebert</bibkey>
+      <video href="2021.acl-long.161.mp4"/>
       <pwccode url="https://github.com/ShannonAI/ChineseBert" additional="true">ShannonAI/ChineseBert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cjrc">CJRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/clue">CLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cmrc">CMRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cmrc-2018">CMRC 2018</pwcdataset>
-      <video href="2021.acl-long.161.mp4"/>
     </paper>
     <paper id="162">
       <title>Weight Distillation: Transferring the Knowledge in Neural Network Parameters</title>
@@ -2618,9 +2621,9 @@
       <url hash="1f634fa8">2021.acl-long.163</url>
       <doi>10.18653/v1/2021.acl-long.163</doi>
       <bibkey>xu-etal-2021-optimizing</bibkey>
+      <video href="2021.acl-long.163.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/reclor">ReClor</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
-      <video href="2021.acl-long.163.mp4"/>
     </paper>
     <paper id="164">
       <title><fixed-case>BERTAC</fixed-case>: Enhancing Transformer-based Language Models with Adversarially Pretrained Convolutional Neural Networks</title>
@@ -2633,13 +2636,13 @@
       <url hash="f9e6ca78">2021.acl-long.164</url>
       <doi>10.18653/v1/2021.acl-long.164</doi>
       <bibkey>oh-etal-2021-bertac</bibkey>
+      <video href="2021.acl-long.164.mp4"/>
       <pwccode url="https://github.com/nict-wisdom/bertac" additional="false">nict-wisdom/bertac</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quasar-1">QUASAR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quasar-t">QUASAR-T</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/searchqa">SearchQA</pwcdataset>
-      <video href="2021.acl-long.164.mp4"/>
     </paper>
     <paper id="165">
       <title><fixed-case>COVID</fixed-case>-Fact: Fact Extraction and Verification of Real-World Claims on <fixed-case>COVID</fixed-case>-19 Pandemic</title>
@@ -2651,6 +2654,7 @@
       <url hash="ecd7d18c">2021.acl-long.165</url>
       <doi>10.18653/v1/2021.acl-long.165</doi>
       <bibkey>saakyan-etal-2021-covid</bibkey>
+      <video href="2021.acl-long.165.mp4"/>
       <pwccode url="https://github.com/asaakyan/covidfact" additional="false">asaakyan/covidfact</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/covid-fact">COVID-Fact</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cord-19">CORD-19</pwcdataset>
@@ -2658,7 +2662,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multifc">MultiFC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scifact">SciFact</pwcdataset>
-      <video href="2021.acl-long.165.mp4"/>
     </paper>
     <paper id="166">
       <title>Explaining Relationships Between Scientific Documents</title>
@@ -2673,8 +2676,8 @@
       <url hash="2f73e655">2021.acl-long.166</url>
       <doi>10.18653/v1/2021.acl-long.166</doi>
       <bibkey>luu-etal-2021-explaining</bibkey>
-      <pwccode url="https://github.com/kel-lu/scigen" additional="false">kel-lu/scigen</pwccode>
       <video href="2021.acl-long.166.mp4"/>
+      <pwccode url="https://github.com/kel-lu/scigen" additional="false">kel-lu/scigen</pwccode>
     </paper>
     <paper id="167">
       <title><fixed-case>I</fixed-case>r<fixed-case>E</fixed-case>ne: Interpretable Energy Prediction for Transformers</title>
@@ -2688,9 +2691,9 @@
       <url hash="439fb105">2021.acl-long.167</url>
       <doi>10.18653/v1/2021.acl-long.167</doi>
       <bibkey>cao-etal-2021-irene</bibkey>
+      <video href="2021.acl-long.167.mp4"/>
       <pwccode url="https://github.com/StonyBrookNLP/irene" additional="false">StonyBrookNLP/irene</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.167.mp4"/>
     </paper>
     <paper id="168">
       <title>Mitigating Bias in Session-based Cyberbullying Detection: A Non-Compromising Approach</title>
@@ -2704,8 +2707,8 @@
       <url hash="d858810d">2021.acl-long.168</url>
       <doi>10.18653/v1/2021.acl-long.168</doi>
       <bibkey>cheng-etal-2021-mitigating</bibkey>
-      <pwccode url="https://github.com/githublucheng/mitigatebiassessioncb" additional="false">githublucheng/mitigatebiassessioncb</pwccode>
       <video href="2021.acl-long.168.mp4"/>
+      <pwccode url="https://github.com/githublucheng/mitigatebiassessioncb" additional="false">githublucheng/mitigatebiassessioncb</pwccode>
     </paper>
     <paper id="169">
       <title><fixed-case>P</fixed-case>lot<fixed-case>C</fixed-case>oder: Hierarchical Decoding for Synthesizing Visualization Code in Programmatic Context</title>
@@ -2718,9 +2721,9 @@
       <url hash="7d3e2d6a">2021.acl-long.169</url>
       <doi>10.18653/v1/2021.acl-long.169</doi>
       <bibkey>chen-etal-2021-plotcoder</bibkey>
+      <video href="2021.acl-long.169.mp4"/>
       <pwccode url="https://github.com/jungyhuk/plotcoder" additional="false">jungyhuk/plotcoder</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/juice">JuICe</pwcdataset>
-      <video href="2021.acl-long.169.mp4"/>
     </paper>
     <paper id="170">
       <title>Changing the World by Changing the Data</title>
@@ -2745,11 +2748,11 @@
       <url hash="aeab902f">2021.acl-long.171</url>
       <doi>10.18653/v1/2021.acl-long.171</doi>
       <bibkey>chen-etal-2021-earlybert</bibkey>
+      <video href="2021.acl-long.171.mp4"/>
       <pwccode url="https://github.com/VITA-Group/EarlyBERT" additional="false">VITA-Group/EarlyBERT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.171.mp4"/>
     </paper>
     <paper id="172">
       <title>On the Effectiveness of Adapter-based Tuning for Pretrained Language Model Adaptation</title>
@@ -2767,9 +2770,9 @@
       <url hash="b6608d07">2021.acl-long.172</url>
       <doi>10.18653/v1/2021.acl-long.172</doi>
       <bibkey>he-etal-2021-effectiveness</bibkey>
+      <video href="2021.acl-long.172.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.172.mp4"/>
     </paper>
     <paper id="173">
       <title>Data Augmentation for Text Generation Without Any Augmented Data</title>
@@ -2799,8 +2802,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="38a95f9a">2021.acl-long.174.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.174</doi>
       <bibkey>ou-etal-2021-integrating</bibkey>
-      <pwccode url="https://github.com/J-zin/SNUH" additional="false">J-zin/SNUH</pwccode>
       <video href="2021.acl-long.174.mp4"/>
+      <pwccode url="https://github.com/J-zin/SNUH" additional="false">J-zin/SNUH</pwccode>
     </paper>
     <paper id="175">
       <title><fixed-case>SMURF</fixed-case>: <fixed-case>S</fixed-case>e<fixed-case>M</fixed-case>antic and linguistic <fixed-case>U</fixed-case>nde<fixed-case>R</fixed-case>standing Fusion for Caption Evaluation via Typicality Analysis</title>
@@ -2811,9 +2814,9 @@
       <url hash="ca4bc400">2021.acl-long.175</url>
       <doi>10.18653/v1/2021.acl-long.175</doi>
       <bibkey>feinglass-yang-2021-smurf</bibkey>
+      <video href="2021.acl-long.175.mp4"/>
       <pwccode url="https://github.com/JoshuaFeinglass/SMURF" additional="false">JoshuaFeinglass/SMURF</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
-      <video href="2021.acl-long.175.mp4"/>
     </paper>
     <paper id="176">
       <title><fixed-case>K</fixed-case>aggle<fixed-case>DBQA</fixed-case>: Realistic Evaluation of Text-to-<fixed-case>SQL</fixed-case> Parsers</title>
@@ -2825,9 +2828,9 @@
       <url hash="4bd70890">2021.acl-long.176</url>
       <doi>10.18653/v1/2021.acl-long.176</doi>
       <bibkey>lee-etal-2021-kaggledbqa</bibkey>
+      <video href="2021.acl-long.176.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
-      <video href="2021.acl-long.176.mp4"/>
     </paper>
     <paper id="177">
       <title><fixed-case>QASR</fixed-case>: <fixed-case>QCRI</fixed-case> Aljazeera Speech Resource A Large Scale Annotated <fixed-case>A</fixed-case>rabic Speech Corpus</title>
@@ -2840,8 +2843,8 @@
       <url hash="3a713f82">2021.acl-long.177</url>
       <doi>10.18653/v1/2021.acl-long.177</doi>
       <bibkey>mubarak-etal-2021-qasr</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/voxceleb1">VoxCeleb1</pwcdataset>
       <video href="2021.acl-long.177.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/voxceleb1">VoxCeleb1</pwcdataset>
     </paper>
     <paper id="178">
       <title>An Empirical Study on Hyperparameter Optimization for Fine-Tuning Pre-trained Language Models</title>
@@ -2853,10 +2856,10 @@
       <url hash="df4944a3">2021.acl-long.178</url>
       <doi>10.18653/v1/2021.acl-long.178</doi>
       <bibkey>liu-wang-2021-empirical</bibkey>
+      <video href="2021.acl-long.178.mp4"/>
       <pwccode url="https://github.com/microsoft/FLAML" additional="false">microsoft/FLAML</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.178.mp4"/>
     </paper>
     <paper id="179">
       <title>Better than Average: Paired Evaluation of <fixed-case>NLP</fixed-case> systems</title>
@@ -2869,8 +2872,8 @@
       <url hash="888fdbd8">2021.acl-long.179</url>
       <doi>10.18653/v1/2021.acl-long.179</doi>
       <bibkey>peyrard-etal-2021-better</bibkey>
-      <pwccode url="https://github.com/epfl-dlab/bt-eval" additional="false">epfl-dlab/bt-eval</pwccode>
       <video href="2021.acl-long.179.mp4"/>
+      <pwccode url="https://github.com/epfl-dlab/bt-eval" additional="false">epfl-dlab/bt-eval</pwccode>
     </paper>
     <paper id="180">
       <title>Chase: A Large-Scale and Pragmatic <fixed-case>C</fixed-case>hinese Dataset for Cross-Database Context-Dependent Text-to-<fixed-case>SQL</fixed-case></title>
@@ -2887,10 +2890,10 @@
       <url hash="ef796f8c">2021.acl-long.180</url>
       <doi>10.18653/v1/2021.acl-long.180</doi>
       <bibkey>guo-etal-2021-chase</bibkey>
+      <video href="2021.acl-long.180.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/csqa">CSQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sparc">SParC</pwcdataset>
-      <video href="2021.acl-long.180.mp4"/>
     </paper>
     <paper id="181">
       <title><fixed-case>CLINE</fixed-case>: Contrastive Learning with Semantic Negative Examples for Natural Language Understanding</title>
@@ -2903,12 +2906,12 @@
       <url hash="5fd7d009">2021.acl-long.181</url>
       <doi>10.18653/v1/2021.acl-long.181</doi>
       <bibkey>wang-etal-2021-cline</bibkey>
+      <video href="2021.acl-long.181.mp4"/>
       <pwccode url="https://github.com/kandorm/CLINE" additional="false">kandorm/CLINE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/perspectrum">Perspectrum</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.acl-long.181.mp4"/>
     </paper>
     <paper id="182">
       <title>Tree-Structured Topic Modeling with Nonparametric Neural Variational Inference</title>
@@ -2922,8 +2925,8 @@
       <url hash="80cb4fd5">2021.acl-long.182</url>
       <doi>10.18653/v1/2021.acl-long.182</doi>
       <bibkey>chen-etal-2021-tree</bibkey>
-      <pwccode url="https://github.com/hostnlp/ntsntm" additional="false">hostnlp/ntsntm</pwccode>
       <video href="2021.acl-long.182.mp4"/>
+      <pwccode url="https://github.com/hostnlp/ntsntm" additional="false">hostnlp/ntsntm</pwccode>
     </paper>
     <paper id="183">
       <title><fixed-case>E</fixed-case>x<fixed-case>CAR</fixed-case>: Event Graph Knowledge Enhanced Explainable Causal Reasoning</title>
@@ -2949,9 +2952,9 @@
       <url hash="3d50d455">2021.acl-long.184</url>
       <doi>10.18653/v1/2021.acl-long.184</doi>
       <bibkey>wang-zong-2021-distributed</bibkey>
+      <video href="2021.acl-long.184.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/goemotions">GoEmotions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/isear">ISEAR</pwcdataset>
-      <video href="2021.acl-long.184.mp4"/>
     </paper>
     <paper id="185">
       <title>Style is <fixed-case>NOT</fixed-case> a single variable: Case Studies for Cross-Stylistic Language Understanding</title>
@@ -2963,13 +2966,13 @@
       <attachment type="OptionalSupplementaryMaterial" hash="a1554fb0">2021.acl-long.185.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-long.185</doi>
       <bibkey>kang-hovy-2021-style</bibkey>
+      <video href="2021.acl-long.185.mp4"/>
       <pwccode url="https://github.com/dykang/xslue" additional="false">dykang/xslue</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/emobank">EmoBank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/pastel">PASTEL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sarc">SARC</pwcdataset>
-      <video href="2021.acl-long.185.mp4"/>
     </paper>
     <paper id="186">
       <title><fixed-case>D</fixed-case>yna<fixed-case>S</fixed-case>ent: A Dynamic Benchmark for Sentiment Analysis</title>
@@ -2982,6 +2985,7 @@
       <url hash="5a9a0366">2021.acl-long.186</url>
       <doi>10.18653/v1/2021.acl-long.186</doi>
       <bibkey>potts-etal-2021-dynasent</bibkey>
+      <video href="2021.acl-long.186.mp4"/>
       <pwccode url="https://github.com/cgpotts/dynasent" additional="false">cgpotts/dynasent</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dynasent">DynaSent</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
@@ -2989,7 +2993,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.186.mp4"/>
     </paper>
     <paper id="187">
       <title>A Hierarchical <fixed-case>VAE</fixed-case> for Calibrating Attributes while Generating Text using Normalizing Flow</title>
@@ -3015,10 +3018,10 @@
       <url hash="a747da7d">2021.acl-long.188</url>
       <doi>10.18653/v1/2021.acl-long.188</doi>
       <bibkey>yan-etal-2021-unified</bibkey>
+      <video href="2021.acl-long.188.mp4"/>
       <pwccode url="https://github.com/yhcc/BARTABSA" additional="true">yhcc/BARTABSA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/aste-data-v2">ASTE-Data-V2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semeval-2014-task-4-sub-task-2">SemEval 2014 Task 4 Sub Task 2</pwcdataset>
-      <video href="2021.acl-long.188.mp4"/>
     </paper>
     <paper id="189">
       <title>Discovering Dialogue Slots with Weak Supervision</title>
@@ -3030,9 +3033,9 @@
       <url hash="5da20a65">2021.acl-long.189</url>
       <doi>10.18653/v1/2021.acl-long.189</doi>
       <bibkey>hudecek-etal-2021-discovering</bibkey>
+      <video href="2021.acl-long.189.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
-      <video href="2021.acl-long.189.mp4"/>
     </paper>
     <paper id="190">
       <title>Enhancing the generalization for Intent Classification and Out-of-Domain Detection in <fixed-case>SLU</fixed-case></title>
@@ -3045,8 +3048,8 @@
       <url hash="b9201e1c">2021.acl-long.190</url>
       <doi>10.18653/v1/2021.acl-long.190</doi>
       <bibkey>shen-etal-2021-enhancing</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <video href="2021.acl-long.190.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
     </paper>
     <paper id="191">
       <title><fixed-case>PROTAUGMENT</fixed-case>: Unsupervised diverse short-texts paraphrasing for intent detection meta-learning</title>
@@ -3059,9 +3062,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="73a1d45b">2021.acl-long.191.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.191</doi>
       <bibkey>dopierre-etal-2021-protaugment</bibkey>
+      <video href="2021.acl-long.191.mp4"/>
       <pwccode url="https://github.com/tdopierre/ProtAugment" additional="false">tdopierre/ProtAugment</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dialoglue">DialoGLUE</pwcdataset>
-      <video href="2021.acl-long.191.mp4"/>
     </paper>
     <paper id="192">
       <title>Robustness Testing of Language Understanding in Task-Oriented Dialog</title>
@@ -3079,8 +3082,8 @@
       <url hash="607bda30">2021.acl-long.192</url>
       <doi>10.18653/v1/2021.acl-long.192</doi>
       <bibkey>liu-etal-2021-robustness</bibkey>
-      <pwccode url="https://github.com/thu-coai/ConvLab-2" additional="true">thu-coai/ConvLab-2</pwccode>
       <video href="2021.acl-long.192.mp4"/>
+      <pwccode url="https://github.com/thu-coai/ConvLab-2" additional="true">thu-coai/ConvLab-2</pwccode>
     </paper>
     <paper id="193">
       <title>Comprehensive Study: How the Context Information of Different Granularity Affects Dialogue State Tracking?</title>
@@ -3092,8 +3095,8 @@
       <url hash="eb6fc1bc">2021.acl-long.193</url>
       <doi>10.18653/v1/2021.acl-long.193</doi>
       <bibkey>yang-etal-2021-comprehensive</bibkey>
-      <pwccode url="https://github.com/yangpuhai/Granularity-in-DST" additional="false">yangpuhai/Granularity-in-DST</pwccode>
       <video href="2021.acl-long.193.mp4"/>
+      <pwccode url="https://github.com/yangpuhai/Granularity-in-DST" additional="false">yangpuhai/Granularity-in-DST</pwccode>
     </paper>
     <paper id="194">
       <title><fixed-case>OTT</fixed-case>ers: One-turn Topic Transitions for Open-Domain Dialogue</title>
@@ -3106,12 +3109,12 @@
       <url hash="867c1690">2021.acl-long.194</url>
       <doi>10.18653/v1/2021.acl-long.194</doi>
       <bibkey>sevegnani-etal-2021-otters</bibkey>
+      <video href="2021.acl-long.194.mp4"/>
       <pwccode url="https://github.com/karinseve/OTTers" additional="false">karinseve/OTTers</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/otters">OTTers</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/topical-chat">Topical-Chat</pwcdataset>
-      <video href="2021.acl-long.194.mp4"/>
     </paper>
     <paper id="195">
       <title>Towards Robustness of Text-to-<fixed-case>SQL</fixed-case> Models against Synonym Substitution</title>
@@ -3127,8 +3130,8 @@
       <url hash="c6438a04">2021.acl-long.195</url>
       <doi>10.18653/v1/2021.acl-long.195</doi>
       <bibkey>gan-etal-2021-towards</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
       <video href="2021.acl-long.195.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
     </paper>
     <paper id="196">
       <title><fixed-case>KACE</fixed-case>: Generating Knowledge Aware Contrastive Explanations for Natural Language Inference</title>
@@ -3144,11 +3147,11 @@
       <url hash="86ff58a0">2021.acl-long.196</url>
       <doi>10.18653/v1/2021.acl-long.196</doi>
       <bibkey>chen-etal-2021-kace</bibkey>
+      <video href="2021.acl-long.196.mp4"/>
       <pwccode url="https://github.com/ai4nlp/kace" additional="false">ai4nlp/kace</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/e-snli">e-SNLI</pwcdataset>
-      <video href="2021.acl-long.196.mp4"/>
     </paper>
     <paper id="197">
       <title>Self-Guided Contrastive Learning for <fixed-case>BERT</fixed-case> Sentence Representations</title>
@@ -3160,9 +3163,9 @@
       <url hash="f59d3af4">2021.acl-long.197</url>
       <doi>10.18653/v1/2021.acl-long.197</doi>
       <bibkey>kim-etal-2021-self</bibkey>
+      <video href="2021.acl-long.197.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.acl-long.197.mp4"/>
     </paper>
     <paper id="198">
       <title><fixed-case>LGESQL</fixed-case>: Line Graph Enhanced Text-to-<fixed-case>SQL</fixed-case> Model with Mixed Local and Non-Local Relations</title>
@@ -3177,9 +3180,9 @@
       <url hash="5e48253f">2021.acl-long.198</url>
       <doi>10.18653/v1/2021.acl-long.198</doi>
       <bibkey>cao-etal-2021-lgesql</bibkey>
+      <video href="2021.acl-long.198.mp4"/>
       <pwccode url="https://github.com/rhythmcao/text2sql-lgesql" additional="false">rhythmcao/text2sql-lgesql</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
-      <video href="2021.acl-long.198.mp4"/>
     </paper>
     <paper id="199">
       <title>Multi-stage Pre-training over Simplified Multimodal Pre-training Models</title>
@@ -3191,13 +3194,13 @@
       <url hash="493e9e38">2021.acl-long.199</url>
       <doi>10.18653/v1/2021.acl-long.199</doi>
       <bibkey>liu-etal-2021-multi</bibkey>
+      <video href="2021.acl-long.199.mp4"/>
       <pwccode url="https://github.com/lttsmn/LXMERT-S" additional="false">lttsmn/LXMERT-S</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gqa">GQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
-      <video href="2021.acl-long.199.mp4"/>
     </paper>
     <paper id="200">
       <title>Beyond Sentence-Level End-to-End Speech Translation: Context Helps</title>
@@ -3210,9 +3213,9 @@
       <url hash="5ed54e03">2021.acl-long.200</url>
       <doi>10.18653/v1/2021.acl-long.200</doi>
       <bibkey>zhang-etal-2021-beyond</bibkey>
+      <video href="2021.acl-long.200.mp4"/>
       <pwccode url="https://github.com/bzhangGo/zero" additional="false">bzhangGo/zero</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
-      <video href="2021.acl-long.200.mp4"/>
     </paper>
     <paper id="201">
       <title><fixed-case>L</fixed-case>ayout<fixed-case>LM</fixed-case>v2: Multi-modal Pre-training for Visually-rich Document Understanding</title>
@@ -3233,6 +3236,7 @@
       <url hash="8841b1e8">2021.acl-long.201</url>
       <doi>10.18653/v1/2021.acl-long.201</doi>
       <bibkey>xu-etal-2021-layoutlmv2</bibkey>
+      <video href="2021.acl-long.201.mp4"/>
       <pwccode url="https://github.com/microsoft/unilm" additional="true">microsoft/unilm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cord">CORD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/docvqa">DocVQA</pwcdataset>
@@ -3241,7 +3245,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/rvl-cdip">RVL-CDIP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sroie">SROIE</pwcdataset>
-      <video href="2021.acl-long.201.mp4"/>
     </paper>
     <paper id="202">
       <title><fixed-case>UNIMO</fixed-case>: Towards Unified-Modal Understanding and Generation via Cross-Modal Contrastive Learning</title>
@@ -3258,6 +3261,7 @@
       <url hash="a26bacc9">2021.acl-long.202</url>
       <doi>10.18653/v1/2021.acl-long.202</doi>
       <bibkey>li-etal-2021-unimo</bibkey>
+      <video href="2021.acl-long.202.mp4"/>
       <pwccode url="https://github.com/PaddlePaddle/PaddleNLP/tree/develop/examples/text_generation/unimo-text" additional="true">PaddlePaddle/PaddleNLP</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coqa">CoQA</pwcdataset>
@@ -3265,7 +3269,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/snli-ve">SNLI-VE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
-      <video href="2021.acl-long.202.mp4"/>
     </paper>
     <paper id="203">
       <title>Missing Modality Imagination Network for Emotion Recognition with Uncertain Missing Modalities</title>
@@ -3277,9 +3280,9 @@
       <url hash="02758b18">2021.acl-long.203</url>
       <doi>10.18653/v1/2021.acl-long.203</doi>
       <bibkey>zhao-etal-2021-missing</bibkey>
+      <video href="2021.acl-long.203.mp4"/>
       <pwccode url="https://github.com/aim3-ruc/mmin" additional="false">aim3-ruc/mmin</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
-      <video href="2021.acl-long.203.mp4"/>
     </paper>
     <paper id="204">
       <title>Stacked Acoustic-and-Textual Encoding: Integrating the Pre-trained Models into Speech Translation Encoders</title>
@@ -3296,9 +3299,9 @@
       <url hash="ebf820a7">2021.acl-long.204</url>
       <doi>10.18653/v1/2021.acl-long.204</doi>
       <bibkey>xu-etal-2021-stacked</bibkey>
+      <video href="2021.acl-long.204.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
-      <video href="2021.acl-long.204.mp4"/>
     </paper>
     <paper id="205">
       <title>N-ary Constituent Tree Parsing with Recursive Semi-<fixed-case>M</fixed-case>arkov Model</title>
@@ -3311,9 +3314,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="cf801d87">2021.acl-long.205.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.205</doi>
       <bibkey>xin-etal-2021-n</bibkey>
+      <video href="2021.acl-long.205.mp4"/>
       <pwccode url="https://github.com/NP-NET-research/Recursive-Semi-Markov-Model" additional="false">NP-NET-research/Recursive-Semi-Markov-Model</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.acl-long.205.mp4"/>
     </paper>
     <paper id="206">
       <title>Automated Concatenation of Embeddings for Structured Prediction</title>
@@ -3329,13 +3332,13 @@
       <url hash="8fe6379a">2021.acl-long.206</url>
       <doi>10.18653/v1/2021.acl-long.206</doi>
       <bibkey>wang-etal-2021-automated</bibkey>
+      <video href="2021.acl-long.206.mp4"/>
       <pwccode url="https://github.com/Alibaba-NLP/ACE" additional="false">Alibaba-NLP/ACE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2002">CoNLL 2002</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semeval-2014-task-4-sub-task-2">SemEval 2014 Task 4 Sub Task 2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tweebank">Tweebank</pwcdataset>
-      <video href="2021.acl-long.206.mp4"/>
     </paper>
     <paper id="207">
       <title>Multi-View Cross-Lingual Structured Prediction with Minimum Supervision</title>
@@ -3374,9 +3377,9 @@
       <url hash="35edd1ca">2021.acl-long.209</url>
       <doi>10.18653/v1/2021.acl-long.209</doi>
       <bibkey>yang-etal-2021-neural</bibkey>
+      <video href="2021.acl-long.209.mp4"/>
       <pwccode url="https://github.com/sustcsonglin/TN-PCFG" additional="false">sustcsonglin/TN-PCFG</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.acl-long.209.mp4"/>
     </paper>
     <paper id="210">
       <title>Ruddit: <fixed-case>N</fixed-case>orms of Offensiveness for <fixed-case>E</fixed-case>nglish <fixed-case>R</fixed-case>eddit Comments</title>
@@ -3393,10 +3396,10 @@
       <bibkey>hada-etal-2021-ruddit</bibkey>
       <revision id="1" href="2021.acl-long.210v1" hash="76242902"/>
       <revision id="2" href="2021.acl-long.210v2" hash="480fc3d5" date="2022-01-24">Updated Table 4, Figure 2 &amp; 3.</revision>
+      <video href="2021.acl-long.210.mp4"/>
       <pwccode url="https://github.com/hadarishav/Ruddit" additional="false">hadarishav/Ruddit</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ruddit">Ruddit</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/pushshift-reddit">Pushshift Reddit</pwcdataset>
-      <video href="2021.acl-long.210.mp4"/>
     </paper>
     <paper id="211">
       <title>Towards Quantifiable Dialogue Coherence Evaluation</title>
@@ -3410,10 +3413,10 @@
       <url hash="8a5142c7">2021.acl-long.211</url>
       <doi>10.18653/v1/2021.acl-long.211</doi>
       <bibkey>ye-etal-2021-towards-quantifiable</bibkey>
+      <video href="2021.acl-long.211.mp4"/>
       <pwccode url="https://github.com/James-Yip/QuantiDCE" additional="false">James-Yip/QuantiDCE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/convai2">ConvAI2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog-1">DailyDialog++</pwcdataset>
-      <video href="2021.acl-long.211.mp4"/>
     </paper>
     <paper id="212">
       <title>Assessing the Representations of Idiomaticity in Vector Models with a Noun Compound Dataset Labeled at Type and Token Levels</title>
@@ -3427,9 +3430,9 @@
       <url hash="8bab2310">2021.acl-long.212</url>
       <doi>10.18653/v1/2021.acl-long.212</doi>
       <bibkey>garcia-etal-2021-assessing</bibkey>
+      <video href="2021.acl-long.212.mp4"/>
       <pwccode url="https://github.com/marcospln/nctti" additional="false">marcospln/nctti</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/brwac">BRWAC</pwcdataset>
-      <video href="2021.acl-long.212.mp4"/>
     </paper>
     <paper id="213">
       <title>Factoring Statutory Reasoning as Language Understanding Challenges</title>
@@ -3440,9 +3443,9 @@
       <url hash="94453965">2021.acl-long.213</url>
       <doi>10.18653/v1/2021.acl-long.213</doi>
       <bibkey>holzenberger-van-durme-2021-factoring</bibkey>
+      <video href="2021.acl-long.213.mp4"/>
       <pwccode url="https://github.com/SgfdDttt/sara_v2" additional="false">SgfdDttt/sara_v2</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sara">SARA</pwcdataset>
-      <video href="2021.acl-long.213.mp4"/>
     </paper>
     <paper id="214">
       <title>Evaluating Evaluation Measures for Ordinal Classification and Ordinal Quantification</title>
@@ -3470,8 +3473,8 @@
       <url hash="eac09d1b">2021.acl-long.215</url>
       <doi>10.18653/v1/2021.acl-long.215</doi>
       <bibkey>yao-etal-2021-interpretable</bibkey>
-      <pwccode url="https://github.com/THU-KEG/HIF-KAT" additional="false">THU-KEG/HIF-KAT</pwccode>
       <video href="2021.acl-long.215.mp4"/>
+      <pwccode url="https://github.com/THU-KEG/HIF-KAT" additional="false">THU-KEG/HIF-KAT</pwccode>
     </paper>
     <paper id="216">
       <title>Locate and Label: A Two-stage Identifier for Nested Named Entity Recognition</title>
@@ -3487,13 +3490,13 @@
       <attachment type="OptionalSupplementaryMaterial" hash="c85c2737">2021.acl-long.216.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.216</doi>
       <bibkey>shen-etal-2021-locate</bibkey>
+      <video href="2021.acl-long.216.mp4"/>
       <pwccode url="https://github.com/tricktreat/locate-and-label" additional="false">tricktreat/locate-and-label</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2004">ACE 2004</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2005">ACE 2005</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/genia">GENIA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/weibo-ner">Weibo NER</pwcdataset>
-      <video href="2021.acl-long.216.mp4"/>
     </paper>
     <paper id="217">
       <title><fixed-case>T</fixed-case>ext2<fixed-case>E</fixed-case>vent: Controllable Sequence-to-Structure Generation for End-to-end Event Extraction</title>
@@ -3511,8 +3514,8 @@
       <url hash="5a7ca51e">2021.acl-long.217</url>
       <doi>10.18653/v1/2021.acl-long.217</doi>
       <bibkey>lu-etal-2021-text2event</bibkey>
-      <pwccode url="https://github.com/luyaojie/text2event" additional="false">luyaojie/text2event</pwccode>
       <video href="2021.acl-long.217.mp4"/>
+      <pwccode url="https://github.com/luyaojie/text2event" additional="false">luyaojie/text2event</pwccode>
     </paper>
     <paper id="218">
       <title>A Large-Scale <fixed-case>C</fixed-case>hinese Multimodal <fixed-case>NER</fixed-case> Dataset with Speech Clues</title>
@@ -3526,10 +3529,10 @@
       <url hash="0d266697">2021.acl-long.218</url>
       <doi>10.18653/v1/2021.acl-long.218</doi>
       <bibkey>sui-etal-2021-large</bibkey>
+      <video href="2021.acl-long.218.mp4"/>
       <pwccode url="https://github.com/dianbowork/cnerta" additional="false">dianbowork/cnerta</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/aishell-1">AISHELL-1</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/genia">GENIA</pwcdataset>
-      <video href="2021.acl-long.218.mp4"/>
     </paper>
     <paper id="219">
       <title>A Neural Transition-based Joint Model for Disease Named Entity Recognition and Normalization</title>
@@ -3561,9 +3564,9 @@
       <bibkey>deng-etal-2021-ontoed</bibkey>
       <revision id="1" href="2021.acl-long.220v1" hash="35e310cc"/>
       <revision id="2" href="2021.acl-long.220v2" hash="015c6379" date="2022-04-29">Added detailed description of evaluation metrics.</revision>
+      <video href="2021.acl-long.220.mp4"/>
       <pwccode url="https://github.com/231sm/Reasoning_In_EE" additional="false">231sm/Reasoning_In_EE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/maven">MAVEN</pwcdataset>
-      <video href="2021.acl-long.220.mp4"/>
     </paper>
     <paper id="221">
       <title>Self-Training Sampling with Monolingual Data Uncertainty for Neural Machine Translation</title>
@@ -3578,8 +3581,8 @@
       <url hash="f8f83d78">2021.acl-long.221</url>
       <doi>10.18653/v1/2021.acl-long.221</doi>
       <bibkey>jiao-etal-2021-self</bibkey>
-      <pwccode url="https://github.com/wxjiao/UncSamp" additional="false">wxjiao/UncSamp</pwccode>
       <video href="2021.acl-long.221.mp4"/>
+      <pwccode url="https://github.com/wxjiao/UncSamp" additional="false">wxjiao/UncSamp</pwccode>
     </paper>
     <paper id="222">
       <title>Breaking the Corpus Bottleneck for Context-Aware Neural Machine Translation with Cross-Task Pre-training</title>
@@ -3655,10 +3658,10 @@
       <url hash="fe40d957">2021.acl-long.226</url>
       <doi>10.18653/v1/2021.acl-long.226</doi>
       <bibkey>mao-etal-2021-lightweight</bibkey>
+      <video href="2021.acl-long.226.mp4"/>
       <pwccode url="https://github.com/Mao-KU/lightweight-crosslingual-sent2vec" additional="false">Mao-KU/lightweight-crosslingual-sent2vec</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mldoc">MLDoc</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paracrawl">ParaCrawl</pwcdataset>
-      <video href="2021.acl-long.226.mp4"/>
     </paper>
     <paper id="227">
       <title><fixed-case>ERNIE</fixed-case>-<fixed-case>D</fixed-case>oc: A Retrospective Long-Document Modeling Transformer</title>
@@ -3681,6 +3684,7 @@
       <pwcdataset url="https://paperswithcode.com/dataset/dureader">DuReader</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="228">
       <title>Marginal Utility Diminishes: Exploring the Minimum Knowledge for <fixed-case>BERT</fixed-case> Knowledge Distillation</title>
@@ -3694,10 +3698,10 @@
       <url hash="3cdbd217">2021.acl-long.228</url>
       <doi>10.18653/v1/2021.acl-long.228</doi>
       <bibkey>liu-etal-2021-marginal</bibkey>
+      <video href="2021.acl-long.228.mp4"/>
       <pwccode url="https://github.com/llyx97/Marginal-Utility-Diminishes" additional="false">llyx97/Marginal-Utility-Diminishes</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.228.mp4"/>
     </paper>
     <paper id="229">
       <title>Rational <fixed-case>LAMOL</fixed-case>: A Rationale-based Lifelong Learning Framework</title>
@@ -3711,10 +3715,10 @@
       <url hash="119c03cd">2021.acl-long.229</url>
       <doi>10.18653/v1/2021.acl-long.229</doi>
       <bibkey>kanwatchara-etal-2021-rational</bibkey>
+      <video href="2021.acl-long.229.mp4"/>
       <pwccode url="https://github.com/kanwatchara-k/r_lamol" additional="false">kanwatchara-k/r_lamol</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scifact">SciFact</pwcdataset>
-      <video href="2021.acl-long.229.mp4"/>
     </paper>
     <paper id="230">
       <title><fixed-case>E</fixed-case>ns<fixed-case>LM</fixed-case>: Ensemble Language Model for Data Diversity by Semantic Clustering</title>
@@ -3729,10 +3733,10 @@
       <url hash="5d12dad5">2021.acl-long.230</url>
       <doi>10.18653/v1/2021.acl-long.230</doi>
       <bibkey>duan-etal-2021-enslm</bibkey>
+      <video href="2021.acl-long.230.mp4"/>
       <pwccode url="https://github.com/bochengroup/enslm" additional="false">bochengroup/enslm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.acl-long.230.mp4"/>
     </paper>
     <paper id="231">
       <title><fixed-case>L</fixed-case>ee<fixed-case>BERT</fixed-case>: Learned Early Exit for <fixed-case>BERT</fixed-case> with cross-level optimization</title>
@@ -3742,9 +3746,9 @@
       <url hash="8ab33401">2021.acl-long.231</url>
       <doi>10.18653/v1/2021.acl-long.231</doi>
       <bibkey>zhu-2021-leebert</bibkey>
+      <video href="2021.acl-long.231.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.231.mp4"/>
     </paper>
     <paper id="232">
       <title>Unsupervised Extractive Summarization-Based Representations for Accurate and Explainable Collaborative Filtering</title>
@@ -3755,8 +3759,8 @@
       <url hash="5b0523ee">2021.acl-long.232</url>
       <doi>10.18653/v1/2021.acl-long.232</doi>
       <bibkey>pugoy-kao-2021-unsupervised</bibkey>
-      <pwccode url="https://github.com/reinaldncku/escofilt" additional="false">reinaldncku/escofilt</pwccode>
       <video href="2021.acl-long.232.mp4"/>
+      <pwccode url="https://github.com/reinaldncku/escofilt" additional="false">reinaldncku/escofilt</pwccode>
     </paper>
     <paper id="233">
       <title><fixed-case>PLOME</fixed-case>: Pre-training with Misspelled Knowledge for <fixed-case>C</fixed-case>hinese Spelling Correction</title>
@@ -3770,8 +3774,8 @@
       <url hash="8fdbc86b">2021.acl-long.233</url>
       <doi>10.18653/v1/2021.acl-long.233</doi>
       <bibkey>liu-etal-2021-plome</bibkey>
-      <pwccode url="https://github.com/liushulinle/plome" additional="false">liushulinle/plome</pwccode>
       <video href="2021.acl-long.233.mp4"/>
+      <pwccode url="https://github.com/liushulinle/plome" additional="false">liushulinle/plome</pwccode>
     </paper>
     <paper id="234">
       <title>Competence-based Multimodal Curriculum Learning for Medical Report Generation</title>
@@ -3783,8 +3787,8 @@
       <url hash="708bfda6">2021.acl-long.234</url>
       <doi>10.18653/v1/2021.acl-long.234</doi>
       <bibkey>liu-etal-2021-competence</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/chexpert">CheXpert</pwcdataset>
       <video href="2021.acl-long.234.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/chexpert">CheXpert</pwcdataset>
     </paper>
     <paper id="235">
       <title>Learning Syntactic Dense Embedding with Correlation Graph for Automatic Readability Assessment</title>
@@ -3799,8 +3803,8 @@
       <url hash="49fe5045">2021.acl-long.235</url>
       <doi>10.18653/v1/2021.acl-long.235</doi>
       <bibkey>qiu-etal-2021-learning</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/onestopenglish">OneStopEnglish</pwcdataset>
       <video href="2021.acl-long.235.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/onestopenglish">OneStopEnglish</pwcdataset>
     </paper>
     <paper id="236">
       <title>Meta-<fixed-case>KD</fixed-case>: A Meta Knowledge Distillation Framework for Language Model Compression across Domains</title>
@@ -3815,8 +3819,8 @@
       <url hash="532c9927">2021.acl-long.236</url>
       <doi>10.18653/v1/2021.acl-long.236</doi>
       <bibkey>pan-etal-2021-meta</bibkey>
-      <pwccode url="https://github.com/alibaba/EasyNLP" additional="false">alibaba/EasyNLP</pwccode>
       <video href="2021.acl-long.236.mp4"/>
+      <pwccode url="https://github.com/alibaba/EasyNLP" additional="false">alibaba/EasyNLP</pwccode>
     </paper>
     <paper id="237">
       <title>A Semantic-based Method for Unsupervised Commonsense Question Answering</title>
@@ -3831,11 +3835,11 @@
       <url hash="d1b84122">2021.acl-long.237</url>
       <doi>10.18653/v1/2021.acl-long.237</doi>
       <bibkey>niu-etal-2021-semantic</bibkey>
+      <video href="2021.acl-long.237.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/atomic">ATOMIC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/social-iqa">Social IQA</pwcdataset>
-      <video href="2021.acl-long.237.mp4"/>
     </paper>
     <paper id="238">
       <title><fixed-case>E</fixed-case>xplanations for <fixed-case>C</fixed-case>ommonsense<fixed-case>QA</fixed-case>: <fixed-case>N</fixed-case>ew <fixed-case>D</fixed-case>ataset and <fixed-case>M</fixed-case>odels</title>
@@ -3851,13 +3855,13 @@
       <attachment type="OptionalSupplementaryMaterial" hash="3df935bc">2021.acl-long.238.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.238</doi>
       <bibkey>aggarwal-etal-2021-explanations</bibkey>
+      <video href="2021.acl-long.238.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cos-e">CoS-E</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/commonsenseqa">CommonsenseQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mathqa">MathQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/openbookqa">OpenBookQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qasc">QASC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/worldtree">Worldtree</pwcdataset>
-      <video href="2021.acl-long.238.mp4"/>
     </paper>
     <paper id="239">
       <title>Few-Shot Question Answering by Pretraining Span Selection</title>
@@ -3871,6 +3875,7 @@
       <url hash="88771cad">2021.acl-long.239</url>
       <doi>10.18653/v1/2021.acl-long.239</doi>
       <bibkey>ram-etal-2021-shot</bibkey>
+      <video href="2021.acl-long.239.mp4"/>
       <pwccode url="https://github.com/oriram/splinter" additional="true">oriram/splinter</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bioasq">BioASQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
@@ -3880,7 +3885,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/searchqa">SearchQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.acl-long.239.mp4"/>
     </paper>
     <paper id="240">
       <title><fixed-case>U</fixed-case>nited<fixed-case>QA</fixed-case>: <fixed-case>A</fixed-case> Hybrid Approach for Open Domain Question Answering</title>
@@ -3895,9 +3899,9 @@
       <url hash="d10997f7">2021.acl-long.240</url>
       <doi>10.18653/v1/2021.acl-long.240</doi>
       <bibkey>cheng-etal-2021-unitedqa</bibkey>
+      <video href="2021.acl-long.240.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.acl-long.240.mp4"/>
     </paper>
     <paper id="241">
       <title>Database reasoning over text</title>
@@ -3912,11 +3916,10 @@
       <url hash="3b5304f8">2021.acl-long.241</url>
       <doi>10.18653/v1/2021.acl-long.241</doi>
       <bibkey>thorne-etal-2021-database</bibkey>
+      <video href="2021.acl-long.241.mp4"/>
       <pwccode url="https://github.com/facebookresearch/NeuralDB" additional="false">facebookresearch/NeuralDB</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikinldb">WikiNLDB</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/break">BREAK</pwcdataset>
-      <pwcdataset url="https://paperswithcode.com/dataset/babi-1">bAbI</pwcdataset>
-      <video href="2021.acl-long.241.mp4"/>
     </paper>
     <paper id="242">
       <title><fixed-case>O</fixed-case>nline <fixed-case>L</fixed-case>earning Meets <fixed-case>M</fixed-case>achine <fixed-case>T</fixed-case>ranslation Evaluation: Finding the Best Systems with the Least Human Effort</title>
@@ -3931,8 +3934,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="1bde1221">2021.acl-long.242.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.242</doi>
       <bibkey>mendonca-etal-2021-online</bibkey>
-      <pwccode url="https://github.com/vania-mendonca/MTOL" additional="false">vania-mendonca/MTOL</pwccode>
       <video href="2021.acl-long.242.mp4"/>
+      <pwccode url="https://github.com/vania-mendonca/MTOL" additional="false">vania-mendonca/MTOL</pwccode>
     </paper>
     <paper id="243">
       <title>How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models</title>
@@ -3946,9 +3949,9 @@
       <url hash="4bc23572">2021.acl-long.243</url>
       <doi>10.18653/v1/2021.acl-long.243</doi>
       <bibkey>rust-etal-2021-good</bibkey>
+      <video href="2021.acl-long.243.mp4"/>
       <pwccode url="https://github.com/Adapter-Hub/hgiyt" additional="false">Adapter-Hub/hgiyt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.acl-long.243.mp4"/>
     </paper>
     <paper id="244">
       <title>Evaluating morphological typology in zero-shot cross-lingual transfer</title>
@@ -4001,9 +4004,9 @@
       <video href="https://www.youtube.com/watch?v=xayfVkt7gwo"/>
       <revision id="1" href="2021.acl-long.247v1" hash="98f2cdc4"/>
       <revision id="2" href="2021.acl-long.247v2" hash="cd7efe76" date="2021-08-31">Typo and grammar fixes</revision>
+      <video href="2021.acl-long.247.mp4"/>
       <pwccode url="https://github.com/phze22/Online-Misogyny-in-Danish-Bajer" additional="false">phze22/Online-Misogyny-in-Danish-Bajer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bajer-danish-misogyny">bajer_danish_misogyny</pwcdataset>
-      <video href="2021.acl-long.247.mp4"/>
     </paper>
     <paper id="248">
       <title>Few-<fixed-case>NERD</fixed-case>: A Few-shot Named Entity Recognition Dataset</title>
@@ -4020,10 +4023,10 @@
       <url hash="8215b04e">2021.acl-long.248</url>
       <doi>10.18653/v1/2021.acl-long.248</doi>
       <bibkey>ding-etal-2021-nerd</bibkey>
+      <video href="2021.acl-long.248.mp4"/>
       <pwccode url="https://github.com/thunlp/Few-NERD" additional="true">thunlp/Few-NERD</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/few-nerd">Few-NERD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wnut-2017-emerging-and-rare-entity">WNUT 2017</pwcdataset>
-      <video href="2021.acl-long.248.mp4"/>
     </paper>
     <paper id="249">
       <title><fixed-case>M</fixed-case>ulti<fixed-case>MET</fixed-case>: A Multimodal Dataset for Metaphor Understanding</title>
@@ -4050,10 +4053,10 @@
       <url hash="1db45410">2021.acl-long.250</url>
       <doi>10.18653/v1/2021.acl-long.250</doi>
       <bibkey>fanton-etal-2021-human</bibkey>
+      <video href="2021.acl-long.250.mp4"/>
       <pwccode url="https://github.com/marcoguerini/CONAN" additional="false">marcoguerini/CONAN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hate-speech">Hate Speech</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sbic">SBIC</pwcdataset>
-      <video href="2021.acl-long.250.mp4"/>
     </paper>
     <paper id="251">
       <title>Can Generative Pre-trained Language Models Serve As Knowledge Bases for Closed-book <fixed-case>QA</fixed-case>?</title>
@@ -4066,12 +4069,12 @@
       <attachment type="OptionalSupplementaryMaterial" hash="7eb056af">2021.acl-long.251.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-long.251</doi>
       <bibkey>wang-etal-2021-generative</bibkey>
+      <video href="2021.acl-long.251.mp4"/>
       <pwccode url="https://github.com/wangcunxiang/Can-PLM-Serve-as-KB-for-CBQA" additional="false">wangcunxiang/Can-PLM-Serve-as-KB-for-CBQA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
-      <video href="2021.acl-long.251.mp4"/>
     </paper>
     <paper id="252">
       <title>Joint Models for Answer Verification in Question Answering Systems</title>
@@ -4083,11 +4086,11 @@
       <url hash="0745bfed">2021.acl-long.252</url>
       <doi>10.18653/v1/2021.acl-long.252</doi>
       <bibkey>zhang-etal-2021-joint</bibkey>
+      <video href="2021.acl-long.252.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/asnq">ASNQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/trecqa">TrecQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikiqa">WikiQA</pwcdataset>
-      <video href="2021.acl-long.252.mp4"/>
     </paper>
     <paper id="253">
       <title>Answering Ambiguous Questions through Generative Evidence Fusion and Round-Trip Prediction</title>
@@ -4106,9 +4109,9 @@
       <url hash="b3562fa0">2021.acl-long.253</url>
       <doi>10.18653/v1/2021.acl-long.253</doi>
       <bibkey>gao-etal-2021-answering</bibkey>
+      <video href="2021.acl-long.253.mp4"/>
       <pwccode url="https://github.com/amzn/refuel-open-domain-qa" additional="false">amzn/refuel-open-domain-qa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.acl-long.253.mp4"/>
     </paper>
     <paper id="254">
       <title><fixed-case>TAT</fixed-case>-<fixed-case>QA</fixed-case>: A Question Answering Benchmark on a Hybrid of Tabular and Textual Content in Finance</title>
@@ -4125,10 +4128,9 @@
       <url hash="2742dd4f">2021.acl-long.254</url>
       <doi>10.18653/v1/2021.acl-long.254</doi>
       <bibkey>zhu-etal-2021-tat</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
-      <pwcdataset url="https://paperswithcode.com/dataset/hybridqa">HybridQA</pwcdataset>
-      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <video href="2021.acl-long.254.mp4"/>
+      <pwccode url="https://github.com/NExTplusplus/TAT-QA" additional="false">NExTplusplus/TAT-QA</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/tat-qa">TAT-QA</pwcdataset>
     </paper>
     <paper id="255">
       <title>Modeling Transitions of Focal Entities for Conversational Knowledge Base Question Answering</title>
@@ -4139,10 +4141,10 @@
       <url hash="39660eb5">2021.acl-long.255</url>
       <doi>10.18653/v1/2021.acl-long.255</doi>
       <bibkey>lan-jiang-2021-modeling</bibkey>
+      <video href="2021.acl-long.255.mp4"/>
       <pwccode url="https://github.com/lanyunshi/conversationalkbqa" additional="false">lanyunshi/conversationalkbqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/csqa">CSQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/convquestions">ConvQuestions</pwcdataset>
-      <video href="2021.acl-long.255.mp4"/>
     </paper>
     <paper id="256">
       <title>Evidence-based Factual Error Correction</title>
@@ -4153,9 +4155,9 @@
       <url hash="94652217">2021.acl-long.256</url>
       <doi>10.18653/v1/2021.acl-long.256</doi>
       <bibkey>thorne-vlachos-2021-evidence</bibkey>
+      <video href="2021.acl-long.256.mp4"/>
       <pwccode url="https://github.com/j6mes/2021-acl-factual-error-correction" additional="false">j6mes/2021-acl-factual-error-correction</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
-      <video href="2021.acl-long.256.mp4"/>
     </paper>
     <paper id="257">
       <title>Probabilistic, Structure-Aware Algorithms for Improved Variety, Accuracy, and Coverage of <fixed-case>AMR</fixed-case> Alignments</title>
@@ -4166,8 +4168,8 @@
       <url hash="8ca80cb5">2021.acl-long.257</url>
       <doi>10.18653/v1/2021.acl-long.257</doi>
       <bibkey>blodgett-schneider-2021-probabilistic</bibkey>
-      <pwccode url="https://github.com/ablodge/leamr" additional="false">ablodge/leamr</pwccode>
       <video href="2021.acl-long.257.mp4"/>
+      <pwccode url="https://github.com/ablodge/leamr" additional="false">ablodge/leamr</pwccode>
     </paper>
     <paper id="258">
       <title>Meta-Learning to Compositionally Generalize</title>
@@ -4180,9 +4182,9 @@
       <url hash="da253c3c">2021.acl-long.258</url>
       <doi>10.18653/v1/2021.acl-long.258</doi>
       <bibkey>conklin-etal-2021-meta</bibkey>
+      <video href="2021.acl-long.258.mp4"/>
       <pwccode url="https://github.com/berlino/tensor2struct-public" additional="false">berlino/tensor2struct-public</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
-      <video href="2021.acl-long.258.mp4"/>
     </paper>
     <paper id="259">
       <title>Taming Pre-trained Language Models with N-gram Representations for Low-Resource Domain Adaptation</title>
@@ -4197,10 +4199,10 @@
       <url hash="5a4d872f">2021.acl-long.259</url>
       <doi>10.18653/v1/2021.acl-long.259</doi>
       <bibkey>diao-etal-2021-taming</bibkey>
+      <video href="2021.acl-long.259.mp4"/>
       <pwccode url="https://github.com/shizhediao/t-dna" additional="false">shizhediao/t-dna</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.acl-long.259.mp4"/>
     </paper>
     <paper id="260">
       <title><fixed-case>ERICA</fixed-case>: Improving Entity and Relation Understanding for Pre-trained Language Models via Contrastive Learning</title>
@@ -4218,6 +4220,7 @@
       <url hash="f693e6cb">2021.acl-long.260</url>
       <doi>10.18653/v1/2021.acl-long.260</doi>
       <bibkey>qin-etal-2021-erica</bibkey>
+      <video href="2021.acl-long.260.mp4"/>
       <pwccode url="https://github.com/thunlp/ERICA" additional="false">thunlp/ERICA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/docred">DocRED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrqa-2019">MRQA</pwcdataset>
@@ -4225,7 +4228,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/semeval-2010-task-8">SemEval-2010 Task 8</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihop">WikiHop</pwcdataset>
-      <video href="2021.acl-long.260.mp4"/>
     </paper>
     <paper id="261">
       <title>Position Bias Mitigation: A Knowledge-Aware Graph Model for Emotion Cause Extraction</title>
@@ -4238,9 +4240,9 @@
       <url hash="8b135fc4">2021.acl-long.261</url>
       <doi>10.18653/v1/2021.acl-long.261</doi>
       <bibkey>yan-etal-2021-position</bibkey>
+      <video href="2021.acl-long.261.mp4"/>
       <pwccode url="https://github.com/hanqi-qi/Position-Bias-Mitigation-in-Emotion-Cause-Analysis" additional="false">hanqi-qi/Position-Bias-Mitigation-in-Emotion-Cause-Analysis</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
-      <video href="2021.acl-long.261.mp4"/>
     </paper>
     <paper id="262">
       <title>Every Bite Is an Experience: <fixed-case>K</fixed-case>ey <fixed-case>P</fixed-case>oint <fixed-case>A</fixed-case>nalysis of Business Reviews</title>
@@ -4269,10 +4271,10 @@
       <url hash="85cfe76b">2021.acl-long.263</url>
       <doi>10.18653/v1/2021.acl-long.263</doi>
       <bibkey>barnes-etal-2021-structured</bibkey>
+      <video href="2021.acl-long.263.mp4"/>
       <pwccode url="https://github.com/jerbarnes/sentiment_graphs" additional="true">jerbarnes/sentiment_graphs</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/norec-fine">NoReC_fine</pwcdataset>
-      <video href="2021.acl-long.263.mp4"/>
     </paper>
     <paper id="264">
       <title>Consistency Regularization for Cross-Lingual Fine-Tuning</title>
@@ -4291,6 +4293,7 @@
       <url hash="32e50330">2021.acl-long.264</url>
       <doi>10.18653/v1/2021.acl-long.264</doi>
       <bibkey>zheng-etal-2021-consistency</bibkey>
+      <video href="2021.acl-long.264.mp4"/>
       <pwccode url="https://github.com/bozheng-hit/xTune" additional="false">bozheng-hit/xTune</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
@@ -4298,7 +4301,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/tydiqa-goldp">TyDiQA-GoldP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.acl-long.264.mp4"/>
     </paper>
     <paper id="265">
       <title>Improving Pretrained Cross-Lingual Language Models via Self-Labeled Word Alignment</title>
@@ -4314,13 +4316,13 @@
       <url hash="d9f18d9b">2021.acl-long.265</url>
       <doi>10.18653/v1/2021.acl-long.265</doi>
       <bibkey>chi-etal-2021-improving</bibkey>
+      <video href="2021.acl-long.265.mp4"/>
       <pwccode url="https://github.com/CZWin32768/XLM-Align" additional="false">CZWin32768/XLM-Align</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.acl-long.265.mp4"/>
     </paper>
     <paper id="266">
       <title>Rejuvenating Low-Frequency Words: Making the Most of Parallel Data in Non-Autoregressive Translation</title>
@@ -4335,8 +4337,8 @@
       <url hash="3890cd29">2021.acl-long.266</url>
       <doi>10.18653/v1/2021.acl-long.266</doi>
       <bibkey>ding-etal-2021-rejuvenating</bibkey>
-      <pwccode url="https://github.com/alphadl/rlfw-nat" additional="false">alphadl/rlfw-nat</pwccode>
       <video href="2021.acl-long.266.mp4"/>
+      <pwccode url="https://github.com/alphadl/rlfw-nat" additional="false">alphadl/rlfw-nat</pwccode>
     </paper>
     <paper id="267">
       <title><fixed-case>G</fixed-case>-Transformer for Document-Level Machine Translation</title>
@@ -4350,8 +4352,8 @@
       <url hash="1e1c9f52">2021.acl-long.267</url>
       <doi>10.18653/v1/2021.acl-long.267</doi>
       <bibkey>bao-etal-2021-g</bibkey>
-      <pwccode url="https://github.com/baoguangsheng/g-transformer" additional="false">baoguangsheng/g-transformer</pwccode>
       <video href="2021.acl-long.267.mp4"/>
+      <pwccode url="https://github.com/baoguangsheng/g-transformer" additional="false">baoguangsheng/g-transformer</pwccode>
     </paper>
     <paper id="268">
       <title>Prevent the Language Model from being Overconfident in Neural Machine Translation</title>
@@ -4365,8 +4367,8 @@
       <url hash="9e0f0528">2021.acl-long.268</url>
       <doi>10.18653/v1/2021.acl-long.268</doi>
       <bibkey>miao-etal-2021-prevent</bibkey>
-      <pwccode url="https://github.com/Mlair77/nmt_adequacy" additional="false">Mlair77/nmt_adequacy</pwccode>
       <video href="2021.acl-long.268.mp4"/>
+      <pwccode url="https://github.com/Mlair77/nmt_adequacy" additional="false">Mlair77/nmt_adequacy</pwccode>
     </paper>
     <paper id="269">
       <title>Towards Emotional Support Dialog Systems</title>
@@ -4383,8 +4385,8 @@
       <url hash="1c7831b0">2021.acl-long.269</url>
       <doi>10.18653/v1/2021.acl-long.269</doi>
       <bibkey>liu-etal-2021-towards</bibkey>
-      <pwccode url="https://github.com/thu-coai/Emotional-Support-Conversation" additional="false">thu-coai/Emotional-Support-Conversation</pwccode>
       <video href="2021.acl-long.269.mp4"/>
+      <pwccode url="https://github.com/thu-coai/Emotional-Support-Conversation" additional="false">thu-coai/Emotional-Support-Conversation</pwccode>
     </paper>
     <paper id="270">
       <title>Novel Slot Detection: A Benchmark for Discovering Unknown Slot Types in the Task-Oriented Dialogue System</title>
@@ -4400,10 +4402,10 @@
       <url hash="d00708e8">2021.acl-long.270</url>
       <doi>10.18653/v1/2021.acl-long.270</doi>
       <bibkey>wu-etal-2021-novel</bibkey>
+      <video href="2021.acl-long.270.mp4"/>
       <pwccode url="https://github.com/ChestnutWYN/ACL2021-Novel-Slot-Detection" additional="false">ChestnutWYN/ACL2021-Novel-Slot-Detection</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.acl-long.270.mp4"/>
     </paper>
     <paper id="271">
       <title><fixed-case>GTM</fixed-case>: A Generative Triple-wise Model for Conversational Question Generation</title>
@@ -4431,9 +4433,9 @@
       <url hash="f56cb0d5">2021.acl-long.272</url>
       <doi>10.18653/v1/2021.acl-long.272</doi>
       <bibkey>wang-etal-2021-diversifying</bibkey>
+      <video href="2021.acl-long.272.mp4"/>
       <pwccode url="https://github.com/lemon234071/AdaLabel" additional="false">lemon234071/AdaLabel</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
-      <video href="2021.acl-long.272.mp4"/>
     </paper>
     <paper id="273">
       <title>Out-of-Scope Intent Detection with Self-Supervision and Discriminative Training</title>
@@ -4448,8 +4450,8 @@
       <url hash="f03253dc">2021.acl-long.273</url>
       <doi>10.18653/v1/2021.acl-long.273</doi>
       <bibkey>zhan-etal-2021-scope</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <video href="2021.acl-long.273.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
     </paper>
     <paper id="274">
       <title>Document-level Event Extraction via Heterogeneous Graph-based Interaction Model with a Tracker</title>
@@ -4462,8 +4464,8 @@
       <url hash="991d2452">2021.acl-long.274</url>
       <doi>10.18653/v1/2021.acl-long.274</doi>
       <bibkey>xu-etal-2021-document</bibkey>
-      <pwccode url="https://github.com/RunxinXu/GIT" additional="true">RunxinXu/GIT</pwccode>
       <video href="2021.acl-long.274.mp4"/>
+      <pwccode url="https://github.com/RunxinXu/GIT" additional="true">RunxinXu/GIT</pwccode>
     </paper>
     <paper id="275">
       <title>Nested Named Entity Recognition via Explicitly Excluding the Influence of the Best Path</title>
@@ -4476,11 +4478,11 @@
       <url hash="72238088">2021.acl-long.275</url>
       <doi>10.18653/v1/2021.acl-long.275</doi>
       <bibkey>wang-etal-2021-nested</bibkey>
+      <video href="2021.acl-long.275.mp4"/>
       <pwccode url="https://github.com/speedcell4/nersted" additional="false">speedcell4/nersted</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2004">ACE 2004</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2005">ACE 2005</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/genia">GENIA</pwcdataset>
-      <video href="2021.acl-long.275.mp4"/>
     </paper>
     <paper id="276">
       <title><fixed-case>L</fixed-case>earn<fixed-case>DA</fixed-case>: Learnable Knowledge-Guided Data Augmentation for Event Causality Identification</title>
@@ -4496,8 +4498,8 @@
       <url hash="e5595068">2021.acl-long.276</url>
       <doi>10.18653/v1/2021.acl-long.276</doi>
       <bibkey>zuo-etal-2021-learnda</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
       <video href="2021.acl-long.276.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
     </paper>
     <paper id="277">
       <title>Revisiting the Negative Data of Distantly Supervised Relation Extraction</title>
@@ -4512,10 +4514,10 @@
       <url hash="2527f93b">2021.acl-long.277</url>
       <doi>10.18653/v1/2021.acl-long.277</doi>
       <bibkey>xie-etal-2021-revisiting</bibkey>
+      <video href="2021.acl-long.277.mp4"/>
       <pwccode url="https://github.com/redreamality/RERE-relation-extraction" additional="false">redreamality/RERE-relation-extraction</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/nyt11-hrl">NYT11-HRL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.acl-long.277.mp4"/>
     </paper>
     <paper id="278">
       <title>Knowing the No-match: Entity Alignment with Dangling Cases</title>
@@ -4527,9 +4529,9 @@
       <url hash="1acb2a6a">2021.acl-long.278</url>
       <doi>10.18653/v1/2021.acl-long.278</doi>
       <bibkey>sun-etal-2021-knowing</bibkey>
+      <video href="2021.acl-long.278.mp4"/>
       <pwccode url="https://github.com/nju-websoft/OpenEA" additional="false">nju-websoft/OpenEA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dbp2-0-zh-en">DBP2.0 zh-en</pwcdataset>
-      <video href="2021.acl-long.278.mp4"/>
     </paper>
     <paper id="279">
       <title>Superbizarre Is Not Superb: Derivational Morphology Improves <fixed-case>BERT</fixed-case>’s Interpretation of Complex Words</title>
@@ -4541,8 +4543,8 @@
       <url hash="2d0f907d">2021.acl-long.279</url>
       <doi>10.18653/v1/2021.acl-long.279</doi>
       <bibkey>hofmann-etal-2021-superbizarre</bibkey>
-      <pwccode url="https://github.com/valentinhofmann/superbizarre" additional="false">valentinhofmann/superbizarre</pwccode>
       <video href="2021.acl-long.279.mp4"/>
+      <pwccode url="https://github.com/valentinhofmann/superbizarre" additional="false">valentinhofmann/superbizarre</pwccode>
     </paper>
     <paper id="280">
       <title><fixed-case>BERT</fixed-case> is to <fixed-case>NLP</fixed-case> what <fixed-case>A</fixed-case>lex<fixed-case>N</fixed-case>et is to <fixed-case>CV</fixed-case>: Can Pre-Trained Language Models Identify Analogies?</title>
@@ -4555,9 +4557,9 @@
       <url hash="943887b8">2021.acl-long.280</url>
       <doi>10.18653/v1/2021.acl-long.280</doi>
       <bibkey>ushio-etal-2021-bert</bibkey>
+      <video href="2021.acl-long.280.mp4"/>
       <pwccode url="https://github.com/asahi417/analogy-language-model" additional="false">asahi417/analogy-language-model</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/analogy-test">Analogy Test</pwcdataset>
-      <video href="2021.acl-long.280.mp4"/>
     </paper>
     <paper id="281">
       <title>Exploring the Representation of Word Meanings in Context: <fixed-case>A</fixed-case> Case Study on Homonymy and Synonymy</title>
@@ -4567,9 +4569,9 @@
       <url hash="b4fee6ab">2021.acl-long.281</url>
       <doi>10.18653/v1/2021.acl-long.281</doi>
       <bibkey>garcia-2021-exploring</bibkey>
+      <video href="2021.acl-long.281.mp4"/>
       <pwccode url="https://github.com/marcospln/homonymy_acl21" additional="false">marcospln/homonymy_acl21</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
-      <video href="2021.acl-long.281.mp4"/>
     </paper>
     <paper id="282">
       <title>Measuring Fine-Grained Domain Relevance of Terms: A Hierarchical Core-Fringe Approach</title>
@@ -4582,8 +4584,8 @@
       <url hash="4a84208e">2021.acl-long.282</url>
       <doi>10.18653/v1/2021.acl-long.282</doi>
       <bibkey>huang-etal-2021-measuring</bibkey>
-      <pwccode url="https://github.com/jeffhj/domain-relevance" additional="false">jeffhj/domain-relevance</pwccode>
       <video href="2021.acl-long.282.mp4"/>
+      <pwccode url="https://github.com/jeffhj/domain-relevance" additional="false">jeffhj/domain-relevance</pwccode>
     </paper>
     <paper id="283">
       <title><fixed-case>HERALD</fixed-case>: An Annotation Efficient Method to Detect User Disengagement in Social Conversations</title>
@@ -4595,9 +4597,9 @@
       <url hash="9d31f603">2021.acl-long.283</url>
       <doi>10.18653/v1/2021.acl-long.283</doi>
       <bibkey>liang-etal-2021-herald</bibkey>
+      <video href="2021.acl-long.283.mp4"/>
       <pwccode url="https://github.com/Weixin-Liang/HERALD" additional="false">Weixin-Liang/HERALD</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/convai2">ConvAI2</pwcdataset>
-      <video href="2021.acl-long.283.mp4"/>
     </paper>
     <paper id="284">
       <title>Value-Agnostic Conversational Semantic Parsing</title>
@@ -4632,8 +4634,8 @@
       <url hash="9fcd3578">2021.acl-long.285</url>
       <doi>10.18653/v1/2021.acl-long.285</doi>
       <bibkey>gu-etal-2021-mpc</bibkey>
-      <pwccode url="https://github.com/JasonForJoy/MPC-BERT" additional="false">JasonForJoy/MPC-BERT</pwccode>
       <video href="2021.acl-long.285.mp4"/>
+      <pwccode url="https://github.com/JasonForJoy/MPC-BERT" additional="false">JasonForJoy/MPC-BERT</pwccode>
     </paper>
     <paper id="286">
       <title>Best of Both Worlds: Making High Accuracy Non-incremental Transformer-based Disfluency Detection Incremental</title>
@@ -4656,9 +4658,9 @@
       <url hash="c6d07c15">2021.acl-long.287</url>
       <doi>10.18653/v1/2021.acl-long.287</doi>
       <bibkey>kim-etal-2021-neuralwoz</bibkey>
+      <video href="2021.acl-long.287.mp4"/>
       <pwccode url="https://github.com/naver-ai/neuralwoz" additional="false">naver-ai/neuralwoz</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
-      <video href="2021.acl-long.287.mp4"/>
     </paper>
     <paper id="288">
       <title><fixed-case>CDRNN</fixed-case>: Discovering Complex Dynamics in Human Language Processing</title>
@@ -4668,9 +4670,9 @@
       <url hash="956e38f4">2021.acl-long.288</url>
       <doi>10.18653/v1/2021.acl-long.288</doi>
       <bibkey>shain-2021-cdrnn</bibkey>
+      <video href="2021.acl-long.288.mp4"/>
       <pwccode url="https://github.com/coryshain/cdr" additional="false">coryshain/cdr</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-stories">Natural Stories</pwcdataset>
-      <video href="2021.acl-long.288.mp4"/>
     </paper>
     <paper id="289">
       <title>Structural Guidance for Transformer Language Models</title>
@@ -4685,9 +4687,9 @@
       <bibkey>qian-etal-2021-structural</bibkey>
       <revision id="1" href="2021.acl-long.289v1" hash="ce37206e"/>
       <revision id="2" href="2021.acl-long.289v2" hash="650ea895" date="2021-09-21">updated paper according to bug fixes</revision>
+      <video href="2021.acl-long.289.mp4"/>
       <pwccode url="https://github.com/IBM/transformers-struct-guidance" additional="false">IBM/transformers-struct-guidance</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/blimp">BLiMP</pwcdataset>
-      <video href="2021.acl-long.289.mp4"/>
     </paper>
     <paper id="290">
       <title>Surprisal Estimators for Human Reading Times Need Character Models</title>
@@ -4699,10 +4701,10 @@
       <url hash="616d7cb9">2021.acl-long.290</url>
       <doi>10.18653/v1/2021.acl-long.290</doi>
       <bibkey>oh-etal-2021-surprisal</bibkey>
+      <video href="2021.acl-long.290.mp4"/>
       <pwccode url="https://github.com/byungdoh/acl21_semproc" additional="false">byungdoh/acl21_semproc</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-stories">Natural Stories</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.acl-long.290.mp4"/>
     </paper>
     <paper id="291">
       <title><fixed-case>C</fixed-case>og<fixed-case>A</fixed-case>lign: Learning to Align Textual Neural Representations to Cognitive Language Processing Signals</title>
@@ -4713,9 +4715,9 @@
       <url hash="9e2f3e58">2021.acl-long.291</url>
       <doi>10.18653/v1/2021.acl-long.291</doi>
       <bibkey>ren-xiong-2021-cogalign</bibkey>
+      <video href="2021.acl-long.291.mp4"/>
       <pwccode url="https://github.com/tjunlp-lab/CogAlign" additional="false">tjunlp-lab/CogAlign</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.291.mp4"/>
     </paper>
     <paper id="292">
       <title>Self-Attention Networks Can Process Bounded Hierarchical Languages</title>
@@ -4728,9 +4730,10 @@
       <url hash="85ab3892">2021.acl-long.292</url>
       <doi>10.18653/v1/2021.acl-long.292</doi>
       <bibkey>yao-etal-2021-self</bibkey>
+      <video href="2021.acl-long.292.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/dyck-transformer" additional="false">princeton-nlp/dyck-transformer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.acl-long.292.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="293">
       <title><fixed-case>T</fixed-case>ext<fixed-case>SETTR</fixed-case>: Few-Shot Text Style Extraction and Tunable Targeted Restyling</title>
@@ -4756,10 +4759,10 @@
       <url hash="e958a229">2021.acl-long.294</url>
       <doi>10.18653/v1/2021.acl-long.294</doi>
       <bibkey>zhu-soricut-2021-h</bibkey>
+      <video href="2021.acl-long.294.mp4"/>
       <pwccode url="" additional="true"/>
       <pwcdataset url="https://paperswithcode.com/dataset/billion-word-benchmark">Billion Word Benchmark</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/lra">LRA</pwcdataset>
-      <video href="2021.acl-long.294.mp4"/>
     </paper>
     <paper id="295">
       <title>Making Pre-trained Language Models Better Few-shot Learners</title>
@@ -4771,6 +4774,7 @@
       <url hash="819c29cd">2021.acl-long.295</url>
       <doi>10.18653/v1/2021.acl-long.295</doi>
       <bibkey>gao-etal-2021-making</bibkey>
+      <video href="2021.acl-long.295.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/LM-BFF" additional="true">princeton-nlp/LM-BFF</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
@@ -4779,7 +4783,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.295.mp4"/>
     </paper>
     <paper id="296">
       <title>A Sweet Rabbit Hole by <fixed-case>DARCY</fixed-case>: Using Honeypots to Detect Universal Trigger’s Adversarial Attacks</title>
@@ -4805,8 +4808,8 @@
       <url hash="e5070e0c">2021.acl-long.297</url>
       <doi>10.18653/v1/2021.acl-long.297</doi>
       <bibkey>wei-etal-2021-towards</bibkey>
-      <pwccode url="https://github.com/weilingwei96/EBGCN" additional="false">weilingwei96/EBGCN</pwccode>
       <video href="2021.acl-long.297.mp4"/>
+      <pwccode url="https://github.com/weilingwei96/EBGCN" additional="false">weilingwei96/EBGCN</pwccode>
     </paper>
     <paper id="298">
       <title>Label-Specific Dual Graph Neural Network for Multi-Label Text Classification</title>
@@ -4819,8 +4822,8 @@
       <url hash="1ba4302d">2021.acl-long.298</url>
       <doi>10.18653/v1/2021.acl-long.298</doi>
       <bibkey>ma-etal-2021-label</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/rcv1">RCV1</pwcdataset>
       <video href="2021.acl-long.298.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/rcv1">RCV1</pwcdataset>
     </paper>
     <paper id="299">
       <title><fixed-case>TAN</fixed-case>-<fixed-case>NTM</fixed-case>: <fixed-case>T</fixed-case>opic Attention Networks for Neural Topic Modeling</title>
@@ -4834,8 +4837,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="843dc95a">2021.acl-long.299.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.299</doi>
       <bibkey>panwar-etal-2021-tan</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <video href="2021.acl-long.299.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
     </paper>
     <paper id="300">
       <title>Cross-language Sentence Selection via Data Augmentation and Rationale Training</title>
@@ -4862,11 +4865,11 @@
       <url hash="7d7f0bca">2021.acl-long.301</url>
       <doi>10.18653/v1/2021.acl-long.301</doi>
       <bibkey>pappas-androutsopoulos-2021-neural</bibkey>
+      <video href="2021.acl-long.301.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bioasq">BioASQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.301.mp4"/>
     </paper>
     <paper id="302">
       <title><fixed-case>W</fixed-case>-<fixed-case>RST</fixed-case>: Towards a Weighted <fixed-case>RST</fixed-case>-style Discourse Framework</title>
@@ -4878,8 +4881,8 @@
       <url hash="1d8628dc">2021.acl-long.302</url>
       <doi>10.18653/v1/2021.acl-long.302</doi>
       <bibkey>huber-etal-2021-w</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <video href="2021.acl-long.302.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
     </paper>
     <paper id="303">
       <title><fixed-case>ABCD</fixed-case>: A Graph Framework to Convert Complex Sentences to a Covering Set of Simple Sentences</title>
@@ -4891,8 +4894,8 @@
       <url hash="97f25d0a">2021.acl-long.303</url>
       <doi>10.18653/v1/2021.acl-long.303</doi>
       <bibkey>gao-etal-2021-abcd</bibkey>
-      <pwccode url="https://github.com/serenayj/ABCD-ACL2021" additional="true">serenayj/ABCD-ACL2021</pwccode>
       <video href="2021.acl-long.303.mp4"/>
+      <pwccode url="https://github.com/serenayj/ABCD-ACL2021" additional="true">serenayj/ABCD-ACL2021</pwccode>
     </paper>
     <paper id="304">
       <title>Which Linguist Invented the Lightbulb? Presupposition Verification for Question-Answering</title>
@@ -4905,11 +4908,11 @@
       <url hash="20d7feac">2021.acl-long.304</url>
       <doi>10.18653/v1/2021.acl-long.304</doi>
       <bibkey>kim-etal-2021-linguist</bibkey>
+      <video href="2021.acl-long.304.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.304.mp4"/>
     </paper>
     <paper id="305">
       <title>Adversarial Learning for Discourse Rhetorical Structure Parsing</title>
@@ -4921,8 +4924,8 @@
       <url hash="70400942">2021.acl-long.305</url>
       <doi>10.18653/v1/2021.acl-long.305</doi>
       <bibkey>zhang-etal-2021-adversarial</bibkey>
-      <pwccode url="https://github.com/nlp-discourse-soochowu/gan_dp" additional="false">nlp-discourse-soochowu/gan_dp</pwccode>
       <video href="2021.acl-long.305.mp4"/>
+      <pwccode url="https://github.com/nlp-discourse-soochowu/gan_dp" additional="false">nlp-discourse-soochowu/gan_dp</pwccode>
     </paper>
     <paper id="306">
       <title>Exploring Discourse Structures for Argument Impact Classification</title>
@@ -4935,8 +4938,8 @@
       <url hash="be64b23b">2021.acl-long.306</url>
       <doi>10.18653/v1/2021.acl-long.306</doi>
       <bibkey>liu-etal-2021-exploring-discourse</bibkey>
-      <pwccode url="https://github.com/HKUST-KnowComp/DisCOC" additional="false">HKUST-KnowComp/DisCOC</pwccode>
       <video href="2021.acl-long.306.mp4"/>
+      <pwccode url="https://github.com/HKUST-KnowComp/DisCOC" additional="false">HKUST-KnowComp/DisCOC</pwccode>
     </paper>
     <paper id="307">
       <title>Point, Disambiguate and Copy: Incorporating Bilingual Dictionaries for Neural Machine Translation</title>
@@ -4970,6 +4973,7 @@
       <url hash="88e30680">2021.acl-long.308</url>
       <doi>10.18653/v1/2021.acl-long.308</doi>
       <bibkey>luo-etal-2021-veco</bibkey>
+      <video href="2021.acl-long.308.mp4"/>
       <pwccode url="https://github.com/alibaba/AliceMind" additional="false">alibaba/AliceMind</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bucc">BUCC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
@@ -4977,7 +4981,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.acl-long.308.mp4"/>
     </paper>
     <paper id="309">
       <title>A unified approach to sentence segmentation of punctuated text in many languages</title>
@@ -4988,10 +4991,10 @@
       <url hash="844b0eea">2021.acl-long.309</url>
       <doi>10.18653/v1/2021.acl-long.309</doi>
       <bibkey>wicks-post-2021-unified</bibkey>
+      <video href="2021.acl-long.309.mp4"/>
       <pwccode url="https://github.com/rewicks/ersatz" additional="false">rewicks/ersatz</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ccmatrix">CCMatrix</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikimatrix">WikiMatrix</pwcdataset>
-      <video href="2021.acl-long.309.mp4"/>
     </paper>
     <paper id="310">
       <title>Towards User-Driven Neural Machine Translation</title>
@@ -5008,8 +5011,8 @@
       <url hash="50cd69f3">2021.acl-long.310</url>
       <doi>10.18653/v1/2021.acl-long.310</doi>
       <bibkey>lin-etal-2021-towards</bibkey>
-      <pwccode url="https://github.com/DeepLearnXMU/User-Driven-NMT" additional="false">DeepLearnXMU/User-Driven-NMT</pwccode>
       <video href="2021.acl-long.310.mp4"/>
+      <pwccode url="https://github.com/DeepLearnXMU/User-Driven-NMT" additional="false">DeepLearnXMU/User-Driven-NMT</pwccode>
     </paper>
     <paper id="311">
       <title>End-to-End Lexically Constrained Machine Translation for Morphologically Rich Languages</title>
@@ -5051,9 +5054,9 @@
       <url hash="e2349369">2021.acl-long.313</url>
       <doi>10.18653/v1/2021.acl-long.313</doi>
       <bibkey>malik-etal-2021-ildc</bibkey>
+      <video href="2021.acl-long.313.mp4"/>
       <pwccode url="https://github.com/Exploration-Lab/CJPE" additional="false">Exploration-Lab/CJPE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ildc">ILDC</pwcdataset>
-      <video href="2021.acl-long.313.mp4"/>
     </paper>
     <paper id="314">
       <title>Supporting Cognitive and Emotional Empathic Writing of Students</title>
@@ -5067,8 +5070,8 @@
       <url hash="15ec47d0">2021.acl-long.314</url>
       <doi>10.18653/v1/2021.acl-long.314</doi>
       <bibkey>wambsganss-etal-2021-supporting</bibkey>
-      <pwccode url="https://github.com/thiemowa/empathy_annotated_peer_reviews" additional="false">thiemowa/empathy_annotated_peer_reviews</pwccode>
       <video href="2021.acl-long.314.mp4"/>
+      <pwccode url="https://github.com/thiemowa/empathy_annotated_peer_reviews" additional="false">thiemowa/empathy_annotated_peer_reviews</pwccode>
     </paper>
     <paper id="315">
       <title>Dual Reader-Parser on Hybrid Textual and Tabular Evidence for Open Domain Question Answering</title>
@@ -5105,10 +5108,10 @@
       <url hash="515fa7af">2021.acl-long.316</url>
       <doi>10.18653/v1/2021.acl-long.316</doi>
       <bibkey>mao-etal-2021-generation</bibkey>
+      <video href="2021.acl-long.316.mp4"/>
       <pwccode url="https://github.com/morningmoni/GAR" additional="false">morningmoni/GAR</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.acl-long.316.mp4"/>
     </paper>
     <paper id="317">
       <title>Check It Again:Progressive Visual Question Answering via Visual Entailment</title>
@@ -5122,9 +5125,9 @@
       <url hash="80f12213">2021.acl-long.317</url>
       <doi>10.18653/v1/2021.acl-long.317</doi>
       <bibkey>si-etal-2021-check</bibkey>
+      <video href="2021.acl-long.317.mp4"/>
       <pwccode url="https://github.com/PhoebusSi/SAR" additional="false">PhoebusSi/SAR</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.acl-long.317.mp4"/>
     </paper>
     <paper id="318">
       <title>A Mutual Information Maximization Approach for the Spurious Solution Problem in Weakly Supervised Question Answering</title>
@@ -5137,13 +5140,13 @@
       <url hash="eaa34f85">2021.acl-long.318</url>
       <doi>10.18653/v1/2021.acl-long.318</doi>
       <bibkey>shao-etal-2021-mutual</bibkey>
+      <video href="2021.acl-long.318.mp4"/>
       <pwccode url="https://github.com/ZhihongShao/MIMAX" additional="false">ZhihongShao/MIMAX</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quasar-1">QUASAR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quasar-t">QUASAR-T</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
-      <video href="2021.acl-long.318.mp4"/>
     </paper>
     <paper id="319">
       <title>Breaking Down Walls of Text: How Can <fixed-case>NLP</fixed-case> Benefit Consumer Privacy?</title>
@@ -5157,8 +5160,8 @@
       <url hash="fd475820">2021.acl-long.319</url>
       <doi>10.18653/v1/2021.acl-long.319</doi>
       <bibkey>ravichander-etal-2021-breaking</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/policyqa">PolicyQA</pwcdataset>
       <video href="2021.acl-long.319.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/policyqa">PolicyQA</pwcdataset>
     </paper>
     <paper id="320">
       <title>Supporting Land Reuse of Former Open Pit Mining Sites using Text Classification and Active Learning</title>
@@ -5226,8 +5229,8 @@
       <url hash="c41d204e">2021.acl-long.323</url>
       <doi>10.18653/v1/2021.acl-long.323</doi>
       <bibkey>lison-etal-2021-anonymisation</bibkey>
-      <pwccode url="https://github.com/ildikopilan/anonymisation_acl2021" additional="false">ildikopilan/anonymisation_acl2021</pwccode>
       <video href="2021.acl-long.323.mp4"/>
+      <pwccode url="https://github.com/ildikopilan/anonymisation_acl2021" additional="false">ildikopilan/anonymisation_acl2021</pwccode>
     </paper>
     <paper id="324">
       <title>End-to-End <fixed-case>AMR</fixed-case> Coreference Resolution</title>
@@ -5240,8 +5243,8 @@
       <url hash="fca8250e">2021.acl-long.324</url>
       <doi>10.18653/v1/2021.acl-long.324</doi>
       <bibkey>fu-etal-2021-end</bibkey>
-      <pwccode url="https://github.com/sean-blank/amrcoref" additional="false">sean-blank/amrcoref</pwccode>
       <video href="2021.acl-long.324.mp4"/>
+      <pwccode url="https://github.com/sean-blank/amrcoref" additional="false">sean-blank/amrcoref</pwccode>
     </paper>
     <paper id="325">
       <title>How is <fixed-case>BERT</fixed-case> surprised? Layerwise detection of linguistic anomalies</title>
@@ -5255,9 +5258,9 @@
       <url hash="d27330e1">2021.acl-long.325</url>
       <doi>10.18653/v1/2021.acl-long.325</doi>
       <bibkey>li-etal-2021-bert</bibkey>
+      <video href="2021.acl-long.325.mp4"/>
       <pwccode url="https://github.com/SPOClab-ca/layerwise-anomaly" additional="false">SPOClab-ca/layerwise-anomaly</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/blimp">BLiMP</pwcdataset>
-      <video href="2021.acl-long.325.mp4"/>
     </paper>
     <paper id="326">
       <title>Psycholinguistic Tripartite Graph Network for Personality Detection</title>
@@ -5270,8 +5273,8 @@
       <url hash="1ce7a5d7">2021.acl-long.326</url>
       <doi>10.18653/v1/2021.acl-long.326</doi>
       <bibkey>yang-etal-2021-psycholinguistic</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/pandora">PANDORA</pwcdataset>
       <video href="2021.acl-long.326.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/pandora">PANDORA</pwcdataset>
     </paper>
     <paper id="327">
       <title>Verb Metaphor Detection via Contextual Relation Learning</title>
@@ -5299,8 +5302,8 @@
       <url hash="04593ab6">2021.acl-long.328</url>
       <doi>10.18653/v1/2021.acl-long.328</doi>
       <bibkey>tang-etal-2021-improving</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
       <video href="2021.acl-long.328.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
     </paper>
     <paper id="329">
       <title>Probing Toxic Content in Large Pre-Trained Language Models</title>
@@ -5314,9 +5317,9 @@
       <url hash="4e0e7c14">2021.acl-long.329</url>
       <doi>10.18653/v1/2021.acl-long.329</doi>
       <bibkey>ousidhoum-etal-2021-probing</bibkey>
+      <video href="2021.acl-long.329.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/atomic">ATOMIC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hate-speech-and-offensive-language">Hate Speech and Offensive Language</pwcdataset>
-      <video href="2021.acl-long.329.mp4"/>
     </paper>
     <paper id="330">
       <title>Societal Biases in Language Generation: Progress and Challenges</title>
@@ -5329,8 +5332,8 @@
       <url hash="b4998f24">2021.acl-long.330</url>
       <doi>10.18653/v1/2021.acl-long.330</doi>
       <bibkey>sheng-etal-2021-societal</bibkey>
-      <pwccode url="https://github.com/ewsheng/decoding-biases" additional="false">ewsheng/decoding-biases</pwccode>
       <video href="2021.acl-long.330.mp4"/>
+      <pwccode url="https://github.com/ewsheng/decoding-biases" additional="false">ewsheng/decoding-biases</pwccode>
     </paper>
     <paper id="331">
       <title>Reservoir Transformers</title>
@@ -5345,9 +5348,9 @@
       <url hash="388e4d5f">2021.acl-long.331</url>
       <doi>10.18653/v1/2021.acl-long.331</doi>
       <bibkey>shen-etal-2021-reservoir</bibkey>
+      <video href="2021.acl-long.331.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.331.mp4"/>
     </paper>
     <paper id="332">
       <title>Subsequence Based Deep Active Learning for Named Entity Recognition</title>
@@ -5359,8 +5362,8 @@
       <url hash="b2f38871">2021.acl-long.332</url>
       <doi>10.18653/v1/2021.acl-long.332</doi>
       <bibkey>radmard-etal-2021-subsequence</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <video href="2021.acl-long.332.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
     </paper>
     <paper id="333">
       <title>Convolutions and Self-Attention: <fixed-case>R</fixed-case>e-interpreting Relative Positions in Pre-trained Language Models</title>
@@ -5373,11 +5376,11 @@
       <url hash="77abf04d">2021.acl-long.333</url>
       <doi>10.18653/v1/2021.acl-long.333</doi>
       <bibkey>chang-etal-2021-convolutions</bibkey>
+      <video href="2021.acl-long.333.mp4"/>
       <pwccode url="https://github.com/mlpc-ucsd/BERT_Convolutions" additional="false">mlpc-ucsd/BERT_Convolutions</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.333.mp4"/>
     </paper>
     <paper id="334">
       <title><fixed-case>B</fixed-case>inary<fixed-case>BERT</fixed-case>: Pushing the Limit of <fixed-case>BERT</fixed-case> Quantization</title>
@@ -5395,11 +5398,11 @@
       <url hash="5d005389">2021.acl-long.334</url>
       <doi>10.18653/v1/2021.acl-long.334</doi>
       <bibkey>bai-etal-2021-binarybert</bibkey>
+      <video href="2021.acl-long.334.mp4"/>
       <pwccode url="https://github.com/huawei-noah/Pretrained-Language-Model" additional="false">huawei-noah/Pretrained-Language-Model</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.334.mp4"/>
     </paper>
     <paper id="335">
       <title>Are Pretrained Convolutions Better than Pretrained Transformers?</title>
@@ -5415,6 +5418,7 @@
       <url hash="aa08393d">2021.acl-long.335</url>
       <doi>10.18653/v1/2021.acl-long.335</doi>
       <bibkey>tay-etal-2021-pretrained</bibkey>
+      <video href="2021.acl-long.335.mp4"/>
       <pwccode url="https://github.com/google-research/google-research/tree/master/pretrained_conv" additional="false">google-research/google-research</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
@@ -5422,7 +5426,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.335.mp4"/>
     </paper>
     <paper id="336">
       <title><fixed-case>P</fixed-case>air<fixed-case>RE</fixed-case>: Knowledge Graph Embeddings via Paired Relation Vectors</title>
@@ -5435,10 +5438,10 @@
       <url hash="300097c2">2021.acl-long.336</url>
       <doi>10.18653/v1/2021.acl-long.336</doi>
       <bibkey>chao-etal-2021-pairre</bibkey>
+      <video href="2021.acl-long.336.mp4"/>
       <pwccode url="https://github.com/alipay/KnowledgeGraphEmbeddingsViaPairedRelationVectors_PairRE" additional="false">alipay/KnowledgeGraphEmbeddingsViaPairedRelationVectors_PairRE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ogb">OGB</pwcdataset>
-      <video href="2021.acl-long.336.mp4"/>
     </paper>
     <paper id="337">
       <title>Hierarchy-aware Label Semantics Matching Network for Hierarchical Text Classification</title>
@@ -5451,10 +5454,10 @@
       <url hash="472a568f">2021.acl-long.337</url>
       <doi>10.18653/v1/2021.acl-long.337</doi>
       <bibkey>chen-etal-2021-hierarchy</bibkey>
+      <video href="2021.acl-long.337.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/eurlex57k">EURLEX57K</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rcv1">RCV1</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/web-of-science-dataset">WOS</pwcdataset>
-      <video href="2021.acl-long.337.mp4"/>
     </paper>
     <paper id="338">
       <title><fixed-case>H</fixed-case>idden<fixed-case>C</fixed-case>ut: Simple Data Augmentation for Natural Language Understanding with Better Generalizability</title>
@@ -5467,11 +5470,11 @@
       <url hash="89b69f4e">2021.acl-long.338</url>
       <doi>10.18653/v1/2021.acl-long.338</doi>
       <bibkey>chen-etal-2021-hiddencut</bibkey>
+      <video href="2021.acl-long.338.mp4"/>
       <pwccode url="https://github.com/GT-SALT/HiddenCut" additional="false">GT-SALT/HiddenCut</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.338.mp4"/>
     </paper>
     <paper id="339">
       <title>Neural Stylistic Response Generation with Disentangled Latent Variables</title>
@@ -5484,8 +5487,8 @@
       <url hash="6168ae0d">2021.acl-long.339</url>
       <doi>10.18653/v1/2021.acl-long.339</doi>
       <bibkey>zhu-etal-2021-neural</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <video href="2021.acl-long.339.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
     </paper>
     <paper id="340">
       <title>Intent Classification and Slot Filling for Privacy Policies</title>
@@ -5501,8 +5504,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="3bfecc80">2021.acl-long.340.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.340</doi>
       <bibkey>ahmad-etal-2021-intent</bibkey>
-      <pwccode url="https://github.com/wasiahmad/PolicyIE" additional="false">wasiahmad/PolicyIE</pwccode>
       <video href="2021.acl-long.340.mp4"/>
+      <pwccode url="https://github.com/wasiahmad/PolicyIE" additional="false">wasiahmad/PolicyIE</pwccode>
     </paper>
     <paper id="341">
       <title><fixed-case>RADDLE</fixed-case>: An Evaluation Benchmark and Analysis Platform for Robust Task-oriented Dialog Systems</title>
@@ -5517,9 +5520,9 @@
       <url hash="83dcd3de">2021.acl-long.341</url>
       <doi>10.18653/v1/2021.acl-long.341</doi>
       <bibkey>peng-etal-2021-raddle</bibkey>
+      <video href="2021.acl-long.341.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
-      <video href="2021.acl-long.341.mp4"/>
     </paper>
     <paper id="342">
       <title>Semantic Representation for Dialogue Modeling</title>
@@ -5532,10 +5535,10 @@
       <url hash="d50781a6">2021.acl-long.342</url>
       <doi>10.18653/v1/2021.acl-long.342</doi>
       <bibkey>bai-etal-2021-semantic</bibkey>
+      <video href="2021.acl-long.342.mp4"/>
       <pwccode url="https://github.com/muyeby/AMR-Dialogue" additional="false">muyeby/AMR-Dialogue</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dialogre">DialogRE</pwcdataset>
-      <video href="2021.acl-long.342.mp4"/>
     </paper>
     <paper id="343">
       <title>A Pre-training Strategy for Zero-Resource Response Selection in Knowledge-Grounded Conversations</title>
@@ -5550,9 +5553,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="5bef37e4">2021.acl-long.343.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-long.343</doi>
       <bibkey>tao-etal-2021-pre</bibkey>
+      <video href="2021.acl-long.343.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cmu-dog">CMU DoG</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.acl-long.343.mp4"/>
     </paper>
     <paper id="344">
       <title>Dependency-driven Relation Extraction with Attentive Graph Convolutional Networks</title>
@@ -5567,9 +5570,9 @@
       <bibkey>tian-etal-2021-dependency</bibkey>
       <revision id="1" href="2021.acl-long.344v1" hash="18c525c0"/>
       <revision id="2" href="2021.acl-long.344v2" hash="161854c0" date="2021-11-09">Updated equation 8 and figure 2.</revision>
+      <video href="2021.acl-long.344.mp4"/>
       <pwccode url="https://github.com/cuhksz-nlp/re-agcn" additional="false">cuhksz-nlp/re-agcn</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/semeval-2010-task-8">SemEval-2010 Task 8</pwcdataset>
-      <video href="2021.acl-long.344.mp4"/>
     </paper>
     <paper id="345">
       <title>Evaluating Entity Disambiguation and the Role of Popularity in Retrieval-Based <fixed-case>NLP</fixed-case></title>
@@ -5583,12 +5586,12 @@
       <url hash="7bbb3dbc">2021.acl-long.345</url>
       <doi>10.18653/v1/2021.acl-long.345</doi>
       <bibkey>chen-etal-2021-evaluating</bibkey>
+      <video href="2021.acl-long.345.mp4"/>
       <pwccode url="https://github.com/anthonywchen/AmbER-Sets" additional="false">anthonywchen/AmbER-Sets</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
-      <video href="2021.acl-long.345.mp4"/>
     </paper>
     <paper id="346">
       <title>Evaluation Examples are not Equally Informative: How should that change <fixed-case>NLP</fixed-case> Leaderboards?</title>
@@ -5604,9 +5607,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="76dcaaec">2021.acl-long.346.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.346</doi>
       <bibkey>rodriguez-etal-2021-evaluation</bibkey>
+      <video href="2021.acl-long.346.mp4"/>
       <pwccode url="https://github.com/jplalor/py-irt" additional="false">jplalor/py-irt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.346.mp4"/>
     </paper>
     <paper id="347">
       <title>Claim Matching Beyond <fixed-case>E</fixed-case>nglish to Scale Global Fact-Checking</title>
@@ -5619,8 +5622,8 @@
       <url hash="9f08f16b">2021.acl-long.347</url>
       <doi>10.18653/v1/2021.acl-long.347</doi>
       <bibkey>kazemi-etal-2021-claim</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <video href="2021.acl-long.347.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
     </paper>
     <paper id="348">
       <title><fixed-case>S</fixed-case>em<fixed-case>F</fixed-case>ace: Pre-training Encoder and Decoder with a Semantic Interface for Neural Machine Translation</title>
@@ -5650,8 +5653,8 @@
       <url hash="e6517239">2021.acl-long.349</url>
       <doi>10.18653/v1/2021.acl-long.349</doi>
       <bibkey>bhattacharyya-etal-2021-energy</bibkey>
-      <pwccode url="https://github.com/rooshenas/ebr_mt" additional="false">rooshenas/ebr_mt</pwccode>
       <video href="2021.acl-long.349.mp4"/>
+      <pwccode url="https://github.com/rooshenas/ebr_mt" additional="false">rooshenas/ebr_mt</pwccode>
     </paper>
     <paper id="350">
       <title>Syntax-augmented Multilingual <fixed-case>BERT</fixed-case> for Cross-lingual Transfer</title>
@@ -5664,12 +5667,12 @@
       <url hash="a035eef3">2021.acl-long.350</url>
       <doi>10.18653/v1/2021.acl-long.350</doi>
       <bibkey>ahmad-etal-2021-syntax</bibkey>
+      <video href="2021.acl-long.350.mp4"/>
       <pwccode url="https://github.com/wasiahmad/Syntax-MBERT" additional="false">wasiahmad/Syntax-MBERT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.acl-long.350.mp4"/>
     </paper>
     <paper id="351">
       <title>How to Adapt Your Pretrained Multilingual Model to 1600 Languages</title>
@@ -5694,10 +5697,10 @@
       <url hash="6ff6c3d6">2021.acl-long.352</url>
       <doi>10.18653/v1/2021.acl-long.352</doi>
       <bibkey>li-etal-2021-weakly</bibkey>
+      <video href="2021.acl-long.352.mp4"/>
       <pwccode url="https://github.com/JiachengLi1995/TALLOR" additional="true">JiachengLi1995/TALLOR</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bc5cdr">BC5CDR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
-      <video href="2021.acl-long.352.mp4"/>
     </paper>
     <paper id="353">
       <title>Prefix-Tuning: Optimizing Continuous Prompts for Generation</title>
@@ -5708,9 +5711,9 @@
       <url hash="005a2ff5">2021.acl-long.353</url>
       <doi>10.18653/v1/2021.acl-long.353</doi>
       <bibkey>li-liang-2021-prefix</bibkey>
+      <video href="2021.acl-long.353.mp4"/>
       <pwccode url="https://github.com/XiangLi1999/PrefixTuning" additional="true">XiangLi1999/PrefixTuning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dart">DART</pwcdataset>
-      <video href="2021.acl-long.353.mp4"/>
     </paper>
     <paper id="354">
       <title><fixed-case>O</fixed-case>ne2<fixed-case>S</fixed-case>et: <fixed-case>G</fixed-case>enerating Diverse Keyphrases as a Set</title>
@@ -5724,9 +5727,9 @@
       <url hash="7964d85b">2021.acl-long.354</url>
       <doi>10.18653/v1/2021.acl-long.354</doi>
       <bibkey>ye-etal-2021-one2set</bibkey>
+      <video href="2021.acl-long.354.mp4"/>
       <pwccode url="https://github.com/jiacheng-ye/kg_one2set" additional="false">jiacheng-ye/kg_one2set</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
-      <video href="2021.acl-long.354.mp4"/>
     </paper>
     <paper id="355">
       <title>Continuous Language Generative Flow</title>
@@ -5739,10 +5742,10 @@
       <url hash="3abd6cc9">2021.acl-long.355</url>
       <doi>10.18653/v1/2021.acl-long.355</doi>
       <bibkey>tang-etal-2021-continuous</bibkey>
+      <video href="2021.acl-long.355.mp4"/>
       <pwccode url="https://github.com/zinengtang/continuousflownlg" additional="false">zinengtang/continuousflownlg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/tvqa">TVQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tvqa-1">TVQA+</pwcdataset>
-      <video href="2021.acl-long.355.mp4"/>
     </paper>
     <paper id="356">
       <title><fixed-case>TWAG</fixed-case>: A Topic-Guided <fixed-case>W</fixed-case>ikipedia Abstract Generator</title>
@@ -5758,11 +5761,11 @@
       <attachment type="OptionalSupplementaryMaterial" hash="d999f823">2021.acl-long.356.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.356</doi>
       <bibkey>zhu-etal-2021-twag</bibkey>
+      <video href="2021.acl-long.356.mp4"/>
       <pwccode url="https://github.com/THU-KEG/TWAG" additional="false">THU-KEG/TWAG</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikicatsum">WikiCatSum</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisum">WikiSum</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikipedia-generation">Wikipedia Generation</pwcdataset>
-      <video href="2021.acl-long.356.mp4"/>
     </paper>
     <paper id="357">
       <title><fixed-case>F</fixed-case>orecast<fixed-case>QA</fixed-case>: A Question Answering Challenge for Event Forecasting with Temporal Text Data</title>
@@ -5778,9 +5781,9 @@
       <url hash="6cf1634f">2021.acl-long.357</url>
       <doi>10.18653/v1/2021.acl-long.357</doi>
       <bibkey>jin-etal-2021-forecastqa</bibkey>
+      <video href="2021.acl-long.357.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/forecastqa">ForecastQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
-      <video href="2021.acl-long.357.mp4"/>
     </paper>
     <paper id="358">
       <title>Recursive Tree-Structured Self-Attention for Answer Sentence Selection</title>
@@ -5792,9 +5795,9 @@
       <url hash="5df73d8b">2021.acl-long.358</url>
       <doi>10.18653/v1/2021.acl-long.358</doi>
       <bibkey>mrini-etal-2021-recursive</bibkey>
+      <video href="2021.acl-long.358.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/asnq">ASNQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikiqa">WikiQA</pwcdataset>
-      <video href="2021.acl-long.358.mp4"/>
     </paper>
     <paper id="359">
       <title>How Knowledge Graph and Attention Help? A Qualitative Analysis into Bag-level Relation Extraction</title>
@@ -5807,9 +5810,9 @@
       <url hash="27db9987">2021.acl-long.359</url>
       <doi>10.18653/v1/2021.acl-long.359</doi>
       <bibkey>hu-etal-2021-knowledge</bibkey>
+      <video href="2021.acl-long.359.mp4"/>
       <pwccode url="https://github.com/zig-kwin-hu/how-KG-ATT-help" additional="false">zig-kwin-hu/how-KG-ATT-help</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
-      <video href="2021.acl-long.359.mp4"/>
     </paper>
     <paper id="360">
       <title>Trigger is Not Sufficient: Exploiting Frame-aware Knowledge for Implicit Event Argument Extraction</title>
@@ -5840,8 +5843,8 @@
       <bibkey>liu-etal-2021-element</bibkey>
       <revision id="1" href="2021.acl-long.361v1" hash="502ea135"/>
       <revision id="2" href="2021.acl-long.361v2" hash="0b78b856" date="2021-08-17">Typo fixes in Abstract, clearer content in Related Work and Conclusion</revision>
-      <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
       <video href="2021.acl-long.361.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
     </paper>
     <paper id="362">
       <title><fixed-case>A</fixed-case>da<fixed-case>T</fixed-case>ag: Multi-Attribute Value Extraction from Product Profiles with Adaptive Decoding</title>
@@ -5882,10 +5885,10 @@
       <url hash="8347957c">2021.acl-long.364</url>
       <doi>10.18653/v1/2021.acl-long.364</doi>
       <bibkey>logan-iv-etal-2021-benchmarking</bibkey>
+      <video href="2021.acl-long.364.mp4"/>
       <pwccode url="https://github.com/rloganiv/streaming-cdc" additional="false">rloganiv/streaming-cdc</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/medmentions">MedMentions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/zeshel">ZESHEL</pwcdataset>
-      <video href="2021.acl-long.364.mp4"/>
     </paper>
     <paper id="365">
       <title>Search from History and Reason for Future: Two-stage Reasoning on Temporal Knowledge Graphs</title>
@@ -5901,8 +5904,8 @@
       <url hash="95c1f489">2021.acl-long.365</url>
       <doi>10.18653/v1/2021.acl-long.365</doi>
       <bibkey>li-etal-2021-search</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
       <video href="2021.acl-long.365.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
     </paper>
     <paper id="366">
       <title>Employing Argumentation Knowledge Graphs for Neural Argument Generation</title>
@@ -5916,8 +5919,8 @@
       <url hash="94509e66">2021.acl-long.366</url>
       <doi>10.18653/v1/2021.acl-long.366</doi>
       <bibkey>al-khatib-etal-2021-employing</bibkey>
-      <pwccode url="https://github.com/webis-de/ACL-21" additional="false">webis-de/ACL-21</pwccode>
       <video href="2021.acl-long.366.mp4"/>
+      <pwccode url="https://github.com/webis-de/ACL-21" additional="false">webis-de/ACL-21</pwccode>
     </paper>
     <paper id="367">
       <title>Learning Span-Level Interactions for Aspect Sentiment Triplet Extraction</title>
@@ -5929,9 +5932,9 @@
       <url hash="efa113e0">2021.acl-long.367</url>
       <doi>10.18653/v1/2021.acl-long.367</doi>
       <bibkey>xu-etal-2021-learning</bibkey>
+      <video href="2021.acl-long.367.mp4"/>
       <pwccode url="https://github.com/chiayewken/Span-ASTE" additional="true">chiayewken/Span-ASTE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/aste-data-v2">ASTE-Data-V2</pwcdataset>
-      <video href="2021.acl-long.367.mp4"/>
     </paper>
     <paper id="368">
       <title>On Compositional Generalization of Neural Machine Translation</title>
@@ -5944,11 +5947,11 @@
       <url hash="a0f73bbe">2021.acl-long.368</url>
       <doi>10.18653/v1/2021.acl-long.368</doi>
       <bibkey>li-etal-2021-compositional</bibkey>
+      <video href="2021.acl-long.368.mp4"/>
       <pwccode url="https://github.com/yafuly/CoGnition" additional="false">yafuly/CoGnition</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
-      <video href="2021.acl-long.368.mp4"/>
     </paper>
     <paper id="369">
       <title>Mask-Align: Self-Supervised Neural Word Alignment</title>
@@ -5960,8 +5963,8 @@
       <url hash="20cd522d">2021.acl-long.369</url>
       <doi>10.18653/v1/2021.acl-long.369</doi>
       <bibkey>chen-etal-2021-mask</bibkey>
-      <pwccode url="https://github.com/THUNLP-MT/Mask-Align" additional="false">THUNLP-MT/Mask-Align</pwccode>
       <video href="2021.acl-long.369.mp4"/>
+      <pwccode url="https://github.com/THUNLP-MT/Mask-Align" additional="false">THUNLP-MT/Mask-Align</pwccode>
     </paper>
     <paper id="370">
       <title><fixed-case>GWLAN</fixed-case>: General Word-Level <fixed-case>A</fixed-case>utocompletio<fixed-case>N</fixed-case> for Computer-Aided Translation</title>
@@ -5987,9 +5990,9 @@
       <url hash="2830f1af">2021.acl-long.371</url>
       <doi>10.18653/v1/2021.acl-long.371</doi>
       <bibkey>zhang-etal-2021-de</bibkey>
+      <video href="2021.acl-long.371.mp4"/>
       <pwccode url="https://github.com/zwkatgithub/DSCAU" additional="false">zwkatgithub/DSCAU</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
-      <video href="2021.acl-long.371.mp4"/>
     </paper>
     <paper id="372">
       <title>A Span-Based Model for Joint Overlapped and Discontinuous Named Entity Recognition</title>
@@ -6002,9 +6005,9 @@
       <url hash="47196f73">2021.acl-long.372</url>
       <doi>10.18653/v1/2021.acl-long.372</doi>
       <bibkey>li-etal-2021-span</bibkey>
+      <video href="2021.acl-long.372.mp4"/>
       <pwccode url="https://github.com/foxlf823/sodner" additional="false">foxlf823/sodner</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/genia">GENIA</pwcdataset>
-      <video href="2021.acl-long.372.mp4"/>
     </paper>
     <paper id="373">
       <title><fixed-case>MLB</fixed-case>i<fixed-case>N</fixed-case>et: A Cross-Sentence Collective Event Detection Network</title>
@@ -6018,9 +6021,9 @@
       <url hash="ad6ce41d">2021.acl-long.373</url>
       <doi>10.18653/v1/2021.acl-long.373</doi>
       <bibkey>lou-etal-2021-mlbinet</bibkey>
+      <video href="2021.acl-long.373.mp4"/>
       <pwccode url="https://github.com/zjunlp/DocED" additional="false">zjunlp/DocED</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2005">ACE 2005</pwcdataset>
-      <video href="2021.acl-long.373.mp4"/>
     </paper>
     <paper id="374">
       <title>Exploiting Document Structures and Cluster Consistencies for Event Coreference Resolution</title>
@@ -6045,8 +6048,8 @@
       <url hash="5bb6acfd">2021.acl-long.375</url>
       <doi>10.18653/v1/2021.acl-long.375</doi>
       <bibkey>tian-etal-2021-stereorel</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
       <video href="2021.acl-long.375.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
     </paper>
     <paper id="376">
       <title>Knowledge-Enriched Event Causality Identification via Latent Structure Induction Networks</title>
@@ -6078,10 +6081,10 @@
       <url hash="ec920c09">2021.acl-long.377</url>
       <doi>10.18653/v1/2021.acl-long.377</doi>
       <bibkey>qi-etal-2021-turn</bibkey>
+      <video href="2021.acl-long.377.mp4"/>
       <pwccode url="https://github.com/thunlp/BkdAtk-LWS" additional="false">thunlp/BkdAtk-LWS</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/olid">OLID</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.377.mp4"/>
     </paper>
     <paper id="378">
       <title>Parameter-Efficient Transfer Learning with Diff Pruning</title>
@@ -6093,11 +6096,11 @@
       <url hash="3bce2d8f">2021.acl-long.378</url>
       <doi>10.18653/v1/2021.acl-long.378</doi>
       <bibkey>guo-etal-2021-parameter</bibkey>
+      <video href="2021.acl-long.378.mp4"/>
       <pwccode url="https://github.com/dguo98/DiffPruning" additional="true">dguo98/DiffPruning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.378.mp4"/>
     </paper>
     <paper id="379">
       <title><fixed-case>R</fixed-case>2<fixed-case>D</fixed-case>2: Recursive Transformer based on Differentiable Tree for Interpretable Hierarchical Language Modeling</title>
@@ -6113,10 +6116,10 @@
       <url hash="eeb368a9">2021.acl-long.379</url>
       <doi>10.18653/v1/2021.acl-long.379</doi>
       <bibkey>hu-etal-2021-r2d2</bibkey>
+      <video href="2021.acl-long.379.mp4"/>
       <pwccode url="https://github.com/alipay/StructuredLM_RTDT" additional="false">alipay/StructuredLM_RTDT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
-      <video href="2021.acl-long.379.mp4"/>
     </paper>
     <paper id="380">
       <title>Risk Minimization for Zero-shot Sequence Labeling</title>
@@ -6144,6 +6147,7 @@
       <url hash="0934b850">2021.acl-long.381</url>
       <doi>10.18653/v1/2021.acl-long.381</doi>
       <bibkey>hambardzumyan-etal-2021-warp</bibkey>
+      <video href="2021.acl-long.381.mp4"/>
       <pwccode url="https://github.com/YerevaNN/warp" additional="false">YerevaNN/warp</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fewglue">FewGlue</pwcdataset>
@@ -6156,7 +6160,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semantic-textual-similarity-2012-2016">Semantic Textual Similarity (2012 - 2016)</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.acl-long.381.mp4"/>
     </paper>
     <paper id="382">
       <title>Lexicon Learning for Few Shot Sequence Modeling</title>
@@ -6167,9 +6170,9 @@
       <url hash="70485231">2021.acl-long.382</url>
       <doi>10.18653/v1/2021.acl-long.382</doi>
       <bibkey>akyurek-andreas-2021-lexicon</bibkey>
+      <video href="2021.acl-long.382.mp4"/>
       <pwccode url="https://github.com/ekinakyurek/lexical" additional="false">ekinakyurek/lexical</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
-      <video href="2021.acl-long.382.mp4"/>
     </paper>
     <paper id="383">
       <title>Personalized Transformer for Explainable Recommendation</title>
@@ -6181,9 +6184,9 @@
       <url hash="29dd307a">2021.acl-long.383</url>
       <doi>10.18653/v1/2021.acl-long.383</doi>
       <bibkey>li-etal-2021-personalized</bibkey>
+      <video href="2021.acl-long.383.mp4"/>
       <pwccode url="https://github.com/lileipisces/PETER" additional="false">lileipisces/PETER</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/100doh">100DOH</pwcdataset>
-      <video href="2021.acl-long.383.mp4"/>
     </paper>
     <paper id="384">
       <title>Generating <fixed-case>SOAP</fixed-case> Notes from Doctor-Patient Conversations Using Modular Summarization Techniques</title>
@@ -6207,8 +6210,8 @@
       <url hash="22d1dd39">2021.acl-long.385</url>
       <doi>10.18653/v1/2021.acl-long.385</doi>
       <bibkey>li-shi-2021-tail</bibkey>
-      <pwccode url="https://github.com/lipiji/TtT" additional="false">lipiji/TtT</pwccode>
       <video href="2021.acl-long.385.mp4"/>
+      <pwccode url="https://github.com/lipiji/TtT" additional="false">lipiji/TtT</pwccode>
     </paper>
     <paper id="386">
       <title>Early Detection of Sexual Predators in Chats</title>
@@ -6220,9 +6223,9 @@
       <url hash="bcd5f00f">2021.acl-long.386</url>
       <doi>10.18653/v1/2021.acl-long.386</doi>
       <bibkey>vogt-etal-2021-early</bibkey>
+      <video href="2021.acl-long.386.mp4"/>
       <pwccode url="https://gitlab.com/early-sexual-predator-detection/eSPD-lab" additional="false">early-sexual-predator-detection/eSPD-lab</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/panc">PANC</pwcdataset>
-      <video href="2021.acl-long.386.mp4"/>
     </paper>
     <paper id="387">
       <title>Writing by Memorizing: Hierarchical Retrieval-based Medical Report Generation</title>
@@ -6235,8 +6238,8 @@
       <url hash="685dd388">2021.acl-long.387</url>
       <doi>10.18653/v1/2021.acl-long.387</doi>
       <bibkey>yang-etal-2021-writing</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mimic-cxr">MIMIC-CXR</pwcdataset>
       <video href="2021.acl-long.387.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mimic-cxr">MIMIC-CXR</pwcdataset>
     </paper>
     <paper id="388">
       <title>Concept-Based Label Embedding via Dynamic Routing for Hierarchical Text Classification</title>
@@ -6251,8 +6254,8 @@
       <url hash="b3366f0c">2021.acl-long.388</url>
       <doi>10.18653/v1/2021.acl-long.388</doi>
       <bibkey>wang-etal-2021-concept</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/web-of-science-dataset">WOS</pwcdataset>
       <video href="2021.acl-long.388.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/web-of-science-dataset">WOS</pwcdataset>
     </paper>
     <paper id="389">
       <title><fixed-case>V</fixed-case>isual<fixed-case>S</fixed-case>parta: An Embarrassingly Simple Approach to Large-scale Text-to-Image Search with Weighted Bag-of-words</title>
@@ -6264,10 +6267,10 @@
       <url hash="6ccc40d2">2021.acl-long.389</url>
       <doi>10.18653/v1/2021.acl-long.389</doi>
       <bibkey>lu-etal-2021-visualsparta</bibkey>
+      <video href="2021.acl-long.389.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coco-captions">COCO Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
-      <video href="2021.acl-long.389.mp4"/>
     </paper>
     <paper id="390">
       <title>Few-Shot Text Ranking with Meta Adapted Synthetic Weak Supervision</title>
@@ -6284,9 +6287,9 @@
       <url hash="1cb203f5">2021.acl-long.390</url>
       <doi>10.18653/v1/2021.acl-long.390</doi>
       <bibkey>sun-etal-2021-shot</bibkey>
+      <video href="2021.acl-long.390.mp4"/>
       <pwccode url="https://github.com/thunlp/MetaAdaptRank" additional="false">thunlp/MetaAdaptRank</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
-      <video href="2021.acl-long.390.mp4"/>
     </paper>
     <paper id="391">
       <title>Semi-Supervised Text Classification with Balanced Deep Representation Distributions</title>
@@ -6298,8 +6301,8 @@
       <url hash="a7e18b40">2021.acl-long.391</url>
       <doi>10.18653/v1/2021.acl-long.391</doi>
       <bibkey>li-etal-2021-semi-supervised</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <video href="2021.acl-long.391.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
     </paper>
     <paper id="392">
       <title>Improving Document Representations by Generating Pseudo Query Embeddings for Dense Retrieval</title>
@@ -6314,8 +6317,8 @@
       <url hash="1fdd5042">2021.acl-long.392</url>
       <doi>10.18653/v1/2021.acl-long.392</doi>
       <bibkey>tang-etal-2021-improving-document</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <video href="2021.acl-long.392.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
     </paper>
     <paper id="393">
       <title><fixed-case>C</fixed-case>on<fixed-case>SERT</fixed-case>: A Contrastive Framework for Self-Supervised Sentence Representation Transfer</title>
@@ -6330,10 +6333,10 @@
       <url hash="5975af2d">2021.acl-long.393</url>
       <doi>10.18653/v1/2021.acl-long.393</doi>
       <bibkey>yan-etal-2021-consert</bibkey>
+      <video href="2021.acl-long.393.mp4"/>
       <pwccode url="https://github.com/yym6472/ConSERT" additional="false">yym6472/ConSERT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.acl-long.393.mp4"/>
     </paper>
     <paper id="394">
       <title>Exploring Dynamic Selection of Branch Expansion Orders for Code Generation</title>
@@ -6350,9 +6353,9 @@
       <url hash="d3b512e8">2021.acl-long.394</url>
       <doi>10.18653/v1/2021.acl-long.394</doi>
       <bibkey>jiang-etal-2021-exploring</bibkey>
+      <video href="2021.acl-long.394.mp4"/>
       <pwccode url="https://github.com/DeepLearnXMU/CG-RL" additional="false">DeepLearnXMU/CG-RL</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conala">CoNaLa</pwcdataset>
-      <video href="2021.acl-long.394.mp4"/>
     </paper>
     <paper id="395">
       <title><fixed-case>COINS</fixed-case>: Dynamically Generating <fixed-case>CO</fixed-case>ntextualized Inference Rules for Narrative Story Completion</title>
@@ -6363,11 +6366,11 @@
       <url hash="7add9028">2021.acl-long.395</url>
       <doi>10.18653/v1/2021.acl-long.395</doi>
       <bibkey>paul-frank-2021-coins</bibkey>
+      <video href="2021.acl-long.395.mp4"/>
       <pwccode url="https://github.com/Heidelberg-NLP/COINS" additional="false">Heidelberg-NLP/COINS</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/atomic">ATOMIC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glucose">GLUCOSE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
-      <video href="2021.acl-long.395.mp4"/>
     </paper>
     <paper id="396">
       <title>Reasoning over Entity-Action-Location Graph for Procedural Text Understanding</title>
@@ -6381,9 +6384,9 @@
       <url hash="7aa8d8b8">2021.acl-long.396</url>
       <doi>10.18653/v1/2021.acl-long.396</doi>
       <bibkey>huang-etal-2021-reasoning</bibkey>
+      <video href="2021.acl-long.396.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/propara">ProPara</pwcdataset>
-      <video href="2021.acl-long.396.mp4"/>
     </paper>
     <paper id="397">
       <title>From Paraphrasing to Semantic Parsing: Unsupervised Semantic Parsing via Synchronous Semantic Decoding</title>
@@ -6412,12 +6415,12 @@
       <url hash="f31e6e39">2021.acl-long.398</url>
       <doi>10.18653/v1/2021.acl-long.398</doi>
       <bibkey>li-zhao-2021-pre</bibkey>
+      <video href="2021.acl-long.398.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.398.mp4"/>
     </paper>
     <paper id="399">
       <title>Structural Pre-training for Dialogue Comprehension</title>
@@ -6428,10 +6431,10 @@
       <url hash="6f9f56aa">2021.acl-long.399</url>
       <doi>10.18653/v1/2021.acl-long.399</doi>
       <bibkey>zhang-zhao-2021-structural</bibkey>
+      <video href="2021.acl-long.399.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/douban">Douban</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/douban-conversation-corpus">Douban Conversation Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mutual">MuTual</pwcdataset>
-      <video href="2021.acl-long.399.mp4"/>
     </paper>
     <paper id="400">
       <title><fixed-case>A</fixed-case>uto<fixed-case>T</fixed-case>iny<fixed-case>BERT</fixed-case>: Automatic Hyper-parameter Optimization for Efficient Pre-trained Language Models</title>
@@ -6446,11 +6449,11 @@
       <url hash="0b8e13f0">2021.acl-long.400</url>
       <doi>10.18653/v1/2021.acl-long.400</doi>
       <bibkey>yin-etal-2021-autotinybert</bibkey>
+      <video href="2021.acl-long.400.mp4"/>
       <pwccode url="https://github.com/huawei-noah/Pretrained-Language-Model" additional="false">huawei-noah/Pretrained-Language-Model</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.400.mp4"/>
     </paper>
     <paper id="401">
       <title>Data Augmentation with Adversarial Training for Cross-Lingual <fixed-case>NLI</fixed-case></title>
@@ -6478,12 +6481,12 @@
       <url hash="a793708a">2021.acl-long.402</url>
       <doi>10.18653/v1/2021.acl-long.402</doi>
       <bibkey>zhang-etal-2021-bootstrapped</bibkey>
+      <video href="2021.acl-long.402.mp4"/>
       <pwccode url="https://github.com/yanzhangnlp/bsl" additional="false">yanzhangnlp/bsl</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.acl-long.402.mp4"/>
     </paper>
     <paper id="403">
       <title>Learning Event Graph Knowledge for Abductive Reasoning</title>
@@ -6497,10 +6500,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="c8899d3f">2021.acl-long.403.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.403</doi>
       <bibkey>du-etal-2021-learning</bibkey>
+      <video href="2021.acl-long.403.mp4"/>
       <pwccode url="https://github.com/sjcfr/ege-roberta" additional="false">sjcfr/ege-roberta</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/timetravel">TimeTravel</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vist">VIST</pwcdataset>
-      <video href="2021.acl-long.403.mp4"/>
     </paper>
     <paper id="404">
       <title>A Cognitive Regularizer for Language Modeling</title>
@@ -6512,9 +6515,10 @@
       <url hash="0d784f34">2021.acl-long.404</url>
       <doi>10.18653/v1/2021.acl-long.404</doi>
       <bibkey>wei-etal-2021-cognitive</bibkey>
+      <video href="2021.acl-long.404.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/wiki-40b">Wiki-40B</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.acl-long.404.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="405">
       <title>Lower Perplexity is Not Always Human-Like</title>
@@ -6531,8 +6535,8 @@
       <bibkey>kuribayashi-etal-2021-lower</bibkey>
       <revision id="1" href="2021.acl-long.405v1" hash="3a3c7bd6"/>
       <revision id="2" href="2021.acl-long.405v2" hash="fae3db42" date="2022-01-24">Corrected data stats in Table 1 and updated metric psychometric predictive power.</revision>
-      <pwccode url="https://github.com/kuribayashi4/surprisal_reading_time_en_ja" additional="false">kuribayashi4/surprisal_reading_time_en_ja</pwccode>
       <video href="2021.acl-long.405.mp4"/>
+      <pwccode url="https://github.com/kuribayashi4/surprisal_reading_time_en_ja" additional="false">kuribayashi4/surprisal_reading_time_en_ja</pwccode>
     </paper>
     <paper id="406">
       <title>Word Sense Disambiguation: Towards Interactive Context Exploitation from Both Word and Sense Perspectives</title>
@@ -6543,9 +6547,9 @@
       <url hash="d5f94f90">2021.acl-long.406</url>
       <doi>10.18653/v1/2021.acl-long.406</doi>
       <bibkey>wang-wang-2021-word</bibkey>
+      <video href="2021.acl-long.406.mp4"/>
       <pwccode url="https://github.com/lwmlyy/sace" additional="false">lwmlyy/sace</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
-      <video href="2021.acl-long.406.mp4"/>
     </paper>
     <paper id="407">
       <title>A Knowledge-Guided Framework for Frame Identification</title>
@@ -6561,8 +6565,8 @@
       <url hash="86ad6a0a">2021.acl-long.407</url>
       <doi>10.18653/v1/2021.acl-long.407</doi>
       <bibkey>su-etal-2021-knowledge</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
       <video href="2021.acl-long.407.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
     </paper>
     <paper id="408">
       <title>Obtaining Better Static Word Embeddings Using Contextual Embedding Models</title>
@@ -6573,10 +6577,10 @@
       <url hash="672f9b18">2021.acl-long.408</url>
       <doi>10.18653/v1/2021.acl-long.408</doi>
       <bibkey>gupta-jaggi-2021-obtaining</bibkey>
+      <video href="2021.acl-long.408.mp4"/>
       <pwccode url="https://github.com/epfml/X2Static" additional="false">epfml/X2Static</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.408.mp4"/>
     </paper>
     <paper id="409">
       <title>Meta-Learning with Variational Semantic Memory for Word Sense Disambiguation</title>
@@ -6590,8 +6594,8 @@
       <url hash="2e9850a9">2021.acl-long.409</url>
       <doi>10.18653/v1/2021.acl-long.409</doi>
       <bibkey>du-etal-2021-meta</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
       <video href="2021.acl-long.409.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
     </paper>
     <paper id="410">
       <title><fixed-case>L</fixed-case>ex<fixed-case>F</fixed-case>it: Lexical Fine-Tuning of Pretrained Language Models</title>
@@ -6618,10 +6622,10 @@
       <url hash="35e8f58a">2021.acl-long.411</url>
       <doi>10.18653/v1/2021.acl-long.411</doi>
       <bibkey>hsu-etal-2021-text</bibkey>
+      <video href="2021.acl-long.411.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ljspeech">LJSpeech</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/places205">Places205</pwcdataset>
-      <video href="2021.acl-long.411.mp4"/>
     </paper>
     <paper id="412">
       <title><fixed-case>CTFN</fixed-case>: Hierarchical Learning for Multimodal Sentiment Analysis Using Coupled-Translation Fusion Network</title>
@@ -6637,8 +6641,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="0a3a599b">2021.acl-long.412.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.412</doi>
       <bibkey>tang-etal-2021-ctfn</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
       <video href="2021.acl-long.412.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
     </paper>
     <paper id="413">
       <title>Positional Artefacts Propagate Through Masked Language Model Embeddings</title>
@@ -6650,10 +6654,10 @@
       <url hash="32ef3ec5">2021.acl-long.413</url>
       <doi>10.18653/v1/2021.acl-long.413</doi>
       <bibkey>luo-etal-2021-positional</bibkey>
+      <video href="2021.acl-long.413.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
-      <video href="2021.acl-long.413.mp4"/>
     </paper>
     <paper id="414">
       <title>Language Model Evaluation Beyond Perplexity</title>
@@ -6679,10 +6683,10 @@
       <url hash="e841b1d3">2021.acl-long.415</url>
       <doi>10.18653/v1/2021.acl-long.415</doi>
       <bibkey>situ-etal-2021-learning</bibkey>
+      <video href="2021.acl-long.415.mp4"/>
       <pwccode url="https://github.com/situsnow/l2e" additional="false">situsnow/l2e</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.415.mp4"/>
     </paper>
     <paper id="416">
       <title><fixed-case>S</fixed-case>tereo<fixed-case>S</fixed-case>et: Measuring stereotypical bias in pretrained language models</title>
@@ -6694,9 +6698,9 @@
       <url hash="a58d3030">2021.acl-long.416</url>
       <doi>10.18653/v1/2021.acl-long.416</doi>
       <bibkey>nadeem-etal-2021-stereoset</bibkey>
+      <video href="2021.acl-long.416.mp4"/>
       <pwccode url="https://github.com/moinnadeem/StereoSet" additional="true">moinnadeem/StereoSet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/stereoset">StereoSet</pwcdataset>
-      <video href="2021.acl-long.416.mp4"/>
     </paper>
     <paper id="417">
       <title>Alignment Rationale for Natural Language Inference</title>
@@ -6710,10 +6714,10 @@
       <url hash="a8958912">2021.acl-long.417</url>
       <doi>10.18653/v1/2021.acl-long.417</doi>
       <bibkey>jiang-etal-2021-alignment</bibkey>
+      <video href="2021.acl-long.417.mp4"/>
       <pwccode url="https://github.com/changmenseng/arec" additional="false">changmenseng/arec</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/e-snli">e-SNLI</pwcdataset>
-      <video href="2021.acl-long.417.mp4"/>
     </paper>
     <paper id="418">
       <title>Enabling Lightweight Fine-tuning for Pre-trained Language Model Compression based on Matrix Product Operators</title>
@@ -6728,10 +6732,10 @@
       <url hash="88ffc7d0">2021.acl-long.418</url>
       <doi>10.18653/v1/2021.acl-long.418</doi>
       <bibkey>liu-etal-2021-enabling</bibkey>
+      <video href="2021.acl-long.418.mp4"/>
       <pwccode url="https://github.com/RUCAIBox/MPOP" additional="false">RUCAIBox/MPOP</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.418.mp4"/>
     </paper>
     <paper id="419">
       <title>On Sample Based Explanation Methods for <fixed-case>NLP</fixed-case>: Faithfulness, Efficiency and Semantic Evaluation</title>
@@ -6747,8 +6751,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="c6a2a399">2021.acl-long.419.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.419</doi>
       <bibkey>zhang-etal-2021-sample</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mams">MAMS</pwcdataset>
       <video href="2021.acl-long.419.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mams">MAMS</pwcdataset>
     </paper>
     <paper id="420">
       <title>Syntax-Enhanced Pre-trained Model</title>
@@ -6767,13 +6771,13 @@
       <url hash="8eece4f2">2021.acl-long.420</url>
       <doi>10.18653/v1/2021.acl-long.420</doi>
       <bibkey>xu-etal-2021-syntax</bibkey>
+      <video href="2021.acl-long.420.mp4"/>
       <pwccode url="https://github.com/Hi-ZenanXu/Syntax-Enhanced_Pre-trained_Model" additional="false">Hi-ZenanXu/Syntax-Enhanced_Pre-trained_Model</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cosmosqa">CosmosQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/open-entity-1">Open Entity</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quasar-1">QUASAR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quasar-t">QUASAR-T</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/searchqa">SearchQA</pwcdataset>
-      <video href="2021.acl-long.420.mp4"/>
     </paper>
     <paper id="421">
       <title>Matching Distributions between Model and Data: Cross-domain Knowledge Distillation for Unsupervised Domain Adaptation</title>
@@ -6801,8 +6805,8 @@
       <url hash="727c5576">2021.acl-long.422</url>
       <doi>10.18653/v1/2021.acl-long.422</doi>
       <bibkey>qian-etal-2021-counterfactual</bibkey>
-      <pwccode url="https://github.com/qianc62/corsair" additional="false">qianc62/corsair</pwccode>
       <video href="2021.acl-long.422.mp4"/>
+      <pwccode url="https://github.com/qianc62/corsair" additional="false">qianc62/corsair</pwccode>
     </paper>
     <paper id="423">
       <title><fixed-case>H</fixed-case>ie<fixed-case>R</fixed-case>ec: Hierarchical User Interest Modeling for Personalized News Recommendation</title>
@@ -6818,8 +6822,8 @@
       <url hash="5c346a48">2021.acl-long.423</url>
       <doi>10.18653/v1/2021.acl-long.423</doi>
       <bibkey>qi-etal-2021-hierec</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
       <video href="2021.acl-long.423.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
     </paper>
     <paper id="424">
       <title><fixed-case>PP</fixed-case>-Rec: News Recommendation with Personalized User Interest and Time-aware News Popularity</title>
@@ -6846,9 +6850,9 @@
       <url hash="181ed04c">2021.acl-long.425</url>
       <doi>10.18653/v1/2021.acl-long.425</doi>
       <bibkey>sheng-etal-2021-article</bibkey>
+      <video href="2021.acl-long.425.mp4"/>
       <pwccode url="https://github.com/ictmcg/mtm" additional="false">ictmcg/mtm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snopes">Snopes</pwcdataset>
-      <video href="2021.acl-long.425.mp4"/>
     </paper>
     <paper id="426">
       <title>Defense against Synonym Substitution-based Adversarial Attacks via <fixed-case>D</fixed-case>irichlet Neighborhood Ensemble</title>
@@ -6862,10 +6866,10 @@
       <url hash="95f64a43">2021.acl-long.426</url>
       <doi>10.18653/v1/2021.acl-long.426</doi>
       <bibkey>zhou-etal-2021-defense</bibkey>
+      <video href="2021.acl-long.426.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.acl-long.426.mp4"/>
     </paper>
     <paper id="427">
       <title>Shortformer: Better Language Modeling using Shorter Inputs</title>
@@ -6877,10 +6881,11 @@
       <url hash="dfe673e5">2021.acl-long.427</url>
       <doi>10.18653/v1/2021.acl-long.427</doi>
       <bibkey>press-etal-2021-shortformer</bibkey>
+      <video href="2021.acl-long.427.mp4"/>
       <pwccode url="https://github.com/ofirpress/shortformer" additional="false">ofirpress/shortformer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.acl-long.427.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="428">
       <title><fixed-case>B</fixed-case>andit<fixed-case>MTL</fixed-case>: Bandit-based Multi-task Learning for Text Classification</title>
@@ -6909,12 +6914,12 @@
       <revision id="1" href="2021.acl-long.429v1" hash="4ac070e5"/>
       <revision id="2" href="2021.acl-long.429v2" hash="36b178df" date="2021-08-12">Fixed typos</revision>
       <revision id="3" href="2021.acl-long.429v3" hash="7037f4d9" date="2021-10-11">Added detailed explanations for the divergence of the SCE and NS loss functions</revision>
+      <video href="2021.acl-long.429.mp4"/>
       <pwccode url="https://github.com/kamigaito/acl2021kge" additional="false">kamigaito/acl2021kge</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fb15k">FB15k</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wn18">WN18</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wn18rr">WN18RR</pwcdataset>
-      <video href="2021.acl-long.429.mp4"/>
     </paper>
     <paper id="430">
       <title>De-Confounded Variational Encoder-Decoder for Logical Table-to-Text Generation</title>
@@ -6942,9 +6947,9 @@
       <url hash="c8440200">2021.acl-long.431</url>
       <doi>10.18653/v1/2021.acl-long.431</doi>
       <bibkey>yang-etal-2021-rethinking</bibkey>
+      <video href="2021.acl-long.431.mp4"/>
       <pwccode url="https://github.com/lancopku/sos" additional="false">lancopku/sos</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.acl-long.431.mp4"/>
     </paper>
     <paper id="432">
       <title>Crowdsourcing Learning as Domain Adaptation: <fixed-case>A</fixed-case> Case Study on Named Entity Recognition</title>
@@ -6958,9 +6963,9 @@
       <url hash="a33caded">2021.acl-long.432</url>
       <doi>10.18653/v1/2021.acl-long.432</doi>
       <bibkey>zhang-etal-2021-crowdsourcing</bibkey>
+      <video href="2021.acl-long.432.mp4"/>
       <pwccode url="https://github.com/izhx/CLasDA" additional="false">izhx/CLasDA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
-      <video href="2021.acl-long.432.mp4"/>
     </paper>
     <paper id="433">
       <title>Exploring Distantly-Labeled Rationales in Neural Network Models</title>
@@ -6973,8 +6978,8 @@
       <url hash="b1ad97ac">2021.acl-long.433</url>
       <doi>10.18653/v1/2021.acl-long.433</doi>
       <bibkey>huang-etal-2021-exploring</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <video href="2021.acl-long.433.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
     </paper>
     <paper id="434">
       <title>Learning to Perturb Word Embeddings for Out-of-distribution <fixed-case>QA</fixed-case></title>
@@ -6987,8 +6992,10 @@
       <url hash="4fab376e">2021.acl-long.434</url>
       <doi>10.18653/v1/2021.acl-long.434</doi>
       <bibkey>lee-etal-2021-learning-perturb</bibkey>
-      <pwccode url="https://github.com/seanie12/SWEP" additional="false">seanie12/SWEP</pwccode>
       <video href="2021.acl-long.434.mp4"/>
+      <pwccode url="https://github.com/seanie12/SWEP" additional="false">seanie12/SWEP</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/bioasq">BioASQ</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
     </paper>
     <paper id="435">
       <title>Maria: A Visual Experience Powered Conversational Agent</title>
@@ -7005,11 +7012,11 @@
       <url hash="0e4b83d2">2021.acl-long.435</url>
       <doi>10.18653/v1/2021.acl-long.435</doi>
       <bibkey>liang-etal-2021-maria</bibkey>
+      <video href="2021.acl-long.435.mp4"/>
       <pwccode url="https://github.com/jokieleung/Maria" additional="false">jokieleung/Maria</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/reddit-conversation-corpus">Reddit Conversation Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
-      <video href="2021.acl-long.435.mp4"/>
     </paper>
     <paper id="436">
       <title>A Human-machine Collaborative Framework for Evaluating Malevolence in Dialogues</title>
@@ -7021,8 +7028,8 @@
       <url hash="46767c5d">2021.acl-long.436</url>
       <doi>10.18653/v1/2021.acl-long.436</doi>
       <bibkey>zhang-etal-2021-human</bibkey>
-      <pwccode url="https://github.com/repozhang/case_hmceval" additional="false">repozhang/case_hmceval</pwccode>
       <video href="2021.acl-long.436.mp4"/>
+      <pwccode url="https://github.com/repozhang/case_hmceval" additional="false">repozhang/case_hmceval</pwccode>
     </paper>
     <paper id="437">
       <title>Generating Relevant and Coherent Dialogue Responses using Self-Separated Conditional Variational <fixed-case>A</fixed-case>uto<fixed-case>E</fixed-case>ncoders</title>
@@ -7036,8 +7043,8 @@
       <url hash="c35fb77e">2021.acl-long.437</url>
       <doi>10.18653/v1/2021.acl-long.437</doi>
       <bibkey>sun-etal-2021-generating</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <video href="2021.acl-long.437.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
     </paper>
     <paper id="438">
       <title>Learning to Ask Conversational Questions by Optimizing <fixed-case>L</fixed-case>evenshtein Distance</title>
@@ -7052,9 +7059,9 @@
       <url hash="c921d19d">2021.acl-long.438</url>
       <doi>10.18653/v1/2021.acl-long.438</doi>
       <bibkey>liu-etal-2021-learning-ask</bibkey>
+      <video href="2021.acl-long.438.mp4"/>
       <pwccode url="https://github.com/LZKSKY/CaSE_RISE" additional="false">LZKSKY/CaSE_RISE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/canard">CANARD</pwcdataset>
-      <video href="2021.acl-long.438.mp4"/>
     </paper>
     <paper id="439">
       <title><fixed-case>DVD</fixed-case>: A Diagnostic Dataset for Multi-step Reasoning in Video Grounded Dialogue</title>
@@ -7069,11 +7076,11 @@
       <url hash="0b92f7f5">2021.acl-long.439</url>
       <doi>10.18653/v1/2021.acl-long.439</doi>
       <bibkey>le-etal-2021-dvd</bibkey>
+      <video href="2021.acl-long.439.mp4"/>
       <pwccode url="https://github.com/facebookresearch/DVDialogues" additional="false">facebookresearch/DVDialogues</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cater">CATER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/clevr">CLEVR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/shapes-1">SHAPES</pwcdataset>
-      <video href="2021.acl-long.439.mp4"/>
     </paper>
     <paper id="440">
       <title><fixed-case>MMGCN</fixed-case>: Multimodal Fusion via Deep Graph Convolution Network for Emotion Recognition in Conversation</title>
@@ -7086,10 +7093,10 @@
       <url hash="529ef2ce">2021.acl-long.440</url>
       <doi>10.18653/v1/2021.acl-long.440</doi>
       <bibkey>hu-etal-2021-mmgcn</bibkey>
+      <video href="2021.acl-long.440.mp4"/>
       <pwccode url="https://github.com/hujingwen6666/MMGCN" additional="false">hujingwen6666/MMGCN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
-      <video href="2021.acl-long.440.mp4"/>
     </paper>
     <paper id="441">
       <title><fixed-case>D</fixed-case>yna<fixed-case>E</fixed-case>val: Unifying Turn and Dialogue Level Evaluation</title>
@@ -7105,13 +7112,13 @@
       <url hash="c7bc233c">2021.acl-long.441</url>
       <doi>10.18653/v1/2021.acl-long.441</doi>
       <bibkey>zhang-etal-2021-dynaeval</bibkey>
+      <video href="2021.acl-long.441.mp4"/>
       <pwccode url="https://github.com/e0397123/DynaEval" additional="false">e0397123/DynaEval</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/convai2">ConvAI2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/empatheticdialogues">EmpatheticDialogues</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fed">FED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.acl-long.441.mp4"/>
     </paper>
     <paper id="442">
       <title><fixed-case>C</fixed-case>o<fixed-case>SQA</fixed-case>: 20,000+ Web Queries for Code Search and Question Answering</title>
@@ -7128,12 +7135,12 @@
       <url hash="8d702111">2021.acl-long.442</url>
       <doi>10.18653/v1/2021.acl-long.442</doi>
       <bibkey>huang-etal-2021-cosqa</bibkey>
+      <video href="2021.acl-long.442.mp4"/>
       <pwccode url="https://github.com/Jun-jie-Huang/CoCLR" additional="false">Jun-jie-Huang/CoCLR</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cosqa">CoSQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/codesearchnet">CodeSearchNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/codexglue">CodeXGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/staqc">StaQC</pwcdataset>
-      <video href="2021.acl-long.442.mp4"/>
     </paper>
     <paper id="443">
       <title>Rewriter-Evaluator Architecture for Neural Machine Translation</title>
@@ -7157,11 +7164,11 @@
       <url hash="0e737a84">2021.acl-long.444</url>
       <doi>10.18653/v1/2021.acl-long.444</doi>
       <bibkey>liang-etal-2021-modeling</bibkey>
+      <video href="2021.acl-long.444.mp4"/>
       <pwccode url="https://github.com/XL2248/CPCC" additional="false">XL2248/CPCC</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bmeld">BMELD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/taskmaster-1">Taskmaster-1</pwcdataset>
-      <video href="2021.acl-long.444.mp4"/>
     </paper>
     <paper id="445">
       <title>Importance-based Neuron Allocation for Multilingual Neural Machine Translation</title>
@@ -7174,8 +7181,8 @@
       <url hash="839d5766">2021.acl-long.445</url>
       <doi>10.18653/v1/2021.acl-long.445</doi>
       <bibkey>xie-etal-2021-importance</bibkey>
-      <pwccode url="https://github.com/ictnlp/NA-MNMT" additional="false">ictnlp/NA-MNMT</pwccode>
       <video href="2021.acl-long.445.mp4"/>
+      <pwccode url="https://github.com/ictnlp/NA-MNMT" additional="false">ictnlp/NA-MNMT</pwccode>
     </paper>
     <paper id="446">
       <title>Transfer Learning for Sequence Generation: from Single-source to Multi-source</title>
@@ -7188,8 +7195,8 @@
       <url hash="4766e125">2021.acl-long.446</url>
       <doi>10.18653/v1/2021.acl-long.446</doi>
       <bibkey>huang-etal-2021-transfer</bibkey>
-      <pwccode url="https://github.com/THUNLP-MT/TRICE" additional="false">THUNLP-MT/TRICE</pwccode>
       <video href="2021.acl-long.446.mp4"/>
+      <pwccode url="https://github.com/THUNLP-MT/TRICE" additional="false">THUNLP-MT/TRICE</pwccode>
     </paper>
     <paper id="447">
       <title>A Closer Look at Few-Shot Crosslingual Transfer: The Choice of Shots Matters</title>
@@ -7205,9 +7212,9 @@
       <url hash="ae4b801b">2021.acl-long.447</url>
       <doi>10.18653/v1/2021.acl-long.447</doi>
       <bibkey>zhao-etal-2021-closer</bibkey>
+      <video href="2021.acl-long.447.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/mldoc">MLDoc</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
-      <video href="2021.acl-long.447.mp4"/>
     </paper>
     <paper id="448">
       <title>Coreference Reasoning in Machine Reading Comprehension</title>
@@ -7221,10 +7228,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="96458db9">2021.acl-long.448.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.448</doi>
       <bibkey>wu-etal-2021-coreference</bibkey>
+      <video href="2021.acl-long.448.mp4"/>
       <pwccode url="https://github.com/UKPLab/coref-reasoning-in-qa" additional="false">UKPLab/coref-reasoning-in-qa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/quoref">Quoref</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.448.mp4"/>
     </paper>
     <paper id="449">
       <title>Adapting Unsupervised Syntactic Parsing Methodology for Discourse Dependency Parsing</title>
@@ -7250,8 +7257,8 @@
       <url hash="438e07e9">2021.acl-long.450</url>
       <doi>10.18653/v1/2021.acl-long.450</doi>
       <bibkey>nguyen-etal-2021-conditional</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <video href="2021.acl-long.450.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
     </paper>
     <paper id="451">
       <title>A Unified Generative Framework for Various <fixed-case>NER</fixed-case> Subtasks</title>
@@ -7266,13 +7273,13 @@
       <url hash="990eb4d3">2021.acl-long.451</url>
       <doi>10.18653/v1/2021.acl-long.451</doi>
       <bibkey>yan-etal-2021-unified-generative</bibkey>
+      <video href="2021.acl-long.451.mp4"/>
       <pwccode url="https://github.com/yhcc/BARTNER" additional="false">yhcc/BARTNER</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2004">ACE 2004</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2005">ACE 2005</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/genia">GENIA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ontonotes-5-0">OntoNotes 5.0</pwcdataset>
-      <video href="2021.acl-long.451.mp4"/>
     </paper>
     <paper id="452">
       <title>An In-depth Study on Internal Structure of <fixed-case>C</fixed-case>hinese Words</title>
@@ -7289,8 +7296,8 @@
       <url hash="b69d8ce4">2021.acl-long.452</url>
       <doi>10.18653/v1/2021.acl-long.452</doi>
       <bibkey>gong-etal-2021-depth</bibkey>
-      <pwccode url="https://github.com/SUDA-LA/wist" additional="false">SUDA-LA/wist</pwccode>
       <video href="2021.acl-long.452.mp4"/>
+      <pwccode url="https://github.com/SUDA-LA/wist" additional="false">SUDA-LA/wist</pwccode>
     </paper>
     <paper id="453">
       <title><fixed-case>M</fixed-case>ul<fixed-case>DA</fixed-case>: A Multilingual Data Augmentation Framework for Low-Resource Cross-Lingual <fixed-case>NER</fixed-case></title>
@@ -7320,11 +7327,11 @@
       <bibkey>liu-etal-2021-lexicon</bibkey>
       <revision id="1" href="2021.acl-long.454v1" hash="88836111"/>
       <revision id="2" href="2021.acl-long.454v2" hash="21e090ed" date="2021-12-26">Modified Figures 2 and 4 for clarity</revision>
+      <video href="2021.acl-long.454.mp4"/>
       <pwccode url="https://github.com/liuwei1206/LEBERT" additional="false">liuwei1206/LEBERT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/resume-ner">Resume NER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/weibo-ner">Weibo NER</pwcdataset>
-      <video href="2021.acl-long.454.mp4"/>
     </paper>
     <paper id="455">
       <title>Math Word Problem Solving with Explicit Numerical Values</title>
@@ -7338,9 +7345,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="82377c88">2021.acl-long.455.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.455</doi>
       <bibkey>wu-etal-2021-math</bibkey>
+      <video href="2021.acl-long.455.mp4"/>
       <pwccode url="https://github.com/qinzhuowu/nums2t" additional="false">qinzhuowu/nums2t</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/math23k">Math23K</pwcdataset>
-      <video href="2021.acl-long.455.mp4"/>
     </paper>
     <paper id="456">
       <title>Neural-Symbolic Solver for Math Word Problems with Auxiliary Tasks</title>
@@ -7354,9 +7361,9 @@
       <url hash="a704fa23">2021.acl-long.456</url>
       <doi>10.18653/v1/2021.acl-long.456</doi>
       <bibkey>qin-etal-2021-neural</bibkey>
+      <video href="2021.acl-long.456.mp4"/>
       <pwccode url="https://github.com/QinJinghui/NS-Solver" additional="false">QinJinghui/NS-Solver</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/math23k">Math23K</pwcdataset>
-      <video href="2021.acl-long.456.mp4"/>
     </paper>
     <paper id="457">
       <title><fixed-case>SM</fixed-case>ed<fixed-case>BERT</fixed-case>: A Knowledge-Enhanced Pre-trained Language Model with Structured Semantics for Medical Text Mining</title>
@@ -7371,8 +7378,8 @@
       <url hash="3e11ff08">2021.acl-long.457</url>
       <doi>10.18653/v1/2021.acl-long.457</doi>
       <bibkey>zhang-etal-2021-smedbert</bibkey>
-      <pwccode url="https://github.com/matnlp/smedbert" additional="true">matnlp/smedbert</pwccode>
       <video href="2021.acl-long.457.mp4"/>
+      <pwccode url="https://github.com/matnlp/smedbert" additional="true">matnlp/smedbert</pwccode>
     </paper>
     <paper id="458">
       <title>What is Your Article Based On? Inferring Fine-grained Provenance</title>
@@ -7397,9 +7404,9 @@
       <url hash="f4854846">2021.acl-long.459</url>
       <doi>10.18653/v1/2021.acl-long.459</doi>
       <bibkey>chen-etal-2021-cross-modal</bibkey>
+      <video href="2021.acl-long.459.mp4"/>
       <pwccode url="https://github.com/zhjohnchan/R2GenCMN" additional="false">zhjohnchan/R2GenCMN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/chexpert">CheXpert</pwcdataset>
-      <video href="2021.acl-long.459.mp4"/>
     </paper>
     <paper id="460">
       <title>Controversy and Conformity: from Generalized to Personalized Aggressiveness Detection</title>
@@ -7429,8 +7436,8 @@
       <url hash="65c45f9f">2021.acl-long.461</url>
       <doi>10.18653/v1/2021.acl-long.461</doi>
       <bibkey>liu-etal-2021-multi-perspective</bibkey>
-      <pwccode url="https://github.com/jhliu17/mcr" additional="false">jhliu17/mcr</pwccode>
       <video href="2021.acl-long.461.mp4"/>
+      <pwccode url="https://github.com/jhliu17/mcr" additional="false">jhliu17/mcr</pwccode>
     </paper>
     <paper id="462">
       <title>Instantaneous Grammatical Error Correction with Shallow Aggressive Decoding</title>
@@ -7443,8 +7450,8 @@
       <url hash="efd981f0">2021.acl-long.462</url>
       <doi>10.18653/v1/2021.acl-long.462</doi>
       <bibkey>sun-etal-2021-instantaneous</bibkey>
-      <pwccode url="https://github.com/AutoTemp/Shallow-Aggressive-Decoding" additional="false">AutoTemp/Shallow-Aggressive-Decoding</pwccode>
       <video href="2021.acl-long.462.mp4"/>
+      <pwccode url="https://github.com/AutoTemp/Shallow-Aggressive-Decoding" additional="false">AutoTemp/Shallow-Aggressive-Decoding</pwccode>
     </paper>
     <paper id="463">
       <title>Automatic <fixed-case>ICD</fixed-case> Coding via Interactive Shared Representation Networks with Self-distillation Mechanism</title>
@@ -7493,8 +7500,8 @@
       <url hash="773c49a7">2021.acl-long.465</url>
       <doi>10.18653/v1/2021.acl-long.465</doi>
       <bibkey>cheng-etal-2021-guiding</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <video href="2021.acl-long.465.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
     </paper>
     <paper id="466">
       <title>Improving Encoder by Auxiliary Supervision Tasks for Table-to-Text Generation</title>
@@ -7507,9 +7514,9 @@
       <url hash="d97af630">2021.acl-long.466</url>
       <doi>10.18653/v1/2021.acl-long.466</doi>
       <bibkey>li-etal-2021-improving-encoder</bibkey>
+      <video href="2021.acl-long.466.mp4"/>
       <pwccode url="https://github.com/liang8qi/data2textwithauxiliarysupervision" additional="false">liang8qi/data2textwithauxiliarysupervision</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/rotowire">RotoWire</pwcdataset>
-      <video href="2021.acl-long.466.mp4"/>
     </paper>
     <paper id="467">
       <title><fixed-case>POS</fixed-case>-<fixed-case>C</fixed-case>onstrained <fixed-case>P</fixed-case>arallel <fixed-case>D</fixed-case>ecoding for <fixed-case>N</fixed-case>on-autoregressive <fixed-case>G</fixed-case>eneration</title>
@@ -7523,10 +7530,10 @@
       <url hash="3a03946e">2021.acl-long.467</url>
       <doi>10.18653/v1/2021.acl-long.467</doi>
       <bibkey>yang-etal-2021-pos</bibkey>
+      <video href="2021.acl-long.467.mp4"/>
       <pwccode url="https://github.com/yangkexin/pospd" additional="false">yangkexin/pospd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glge">GLGE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
-      <video href="2021.acl-long.467.mp4"/>
     </paper>
     <paper id="468">
       <title>Bridging Subword Gaps in Pretrain-Finetune Paradigm for Natural Language Generation</title>
@@ -7543,8 +7550,8 @@
       <url hash="42cdfb1d">2021.acl-long.468</url>
       <doi>10.18653/v1/2021.acl-long.468</doi>
       <bibkey>liu-etal-2021-bridging</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <video href="2021.acl-long.468.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
     </paper>
     <paper id="469">
       <title><fixed-case>TGEA</fixed-case>: An Error-Annotated Dataset and Benchmark Tasks for <fixed-case>T</fixed-case>ext<fixed-case>G</fixed-case>eneration from Pretrained Language Models</title>
@@ -7558,6 +7565,7 @@
       <url hash="31c4bbd1">2021.acl-long.469</url>
       <doi>10.18653/v1/2021.acl-long.469</doi>
       <bibkey>he-etal-2021-tgea</bibkey>
+      <video href="2021.acl-long.469.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cosmosqa">CosmosQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fce">FCE</pwcdataset>
@@ -7567,7 +7575,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/social-iqa">Social IQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
-      <video href="2021.acl-long.469.mp4"/>
     </paper>
     <paper id="470">
       <title>Long-Span Summarization via Local Attention and Content Selection</title>
@@ -7578,9 +7585,9 @@
       <url hash="7063efb1">2021.acl-long.470</url>
       <doi>10.18653/v1/2021.acl-long.470</doi>
       <bibkey>manakul-gales-2021-long</bibkey>
+      <video href="2021.acl-long.470.mp4"/>
       <pwccode url="https://github.com/potsawee/longsum0" additional="false">potsawee/longsum0</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
-      <video href="2021.acl-long.470.mp4"/>
     </paper>
     <paper id="471">
       <title><fixed-case>R</fixed-case>ep<fixed-case>S</fixed-case>um: Unsupervised Dialogue Summarization based on Replacement Strategy</title>
@@ -7612,9 +7619,9 @@
       <url hash="8ef86c05">2021.acl-long.472</url>
       <doi>10.18653/v1/2021.acl-long.472</doi>
       <bibkey>wu-etal-2021-bass</bibkey>
+      <video href="2021.acl-long.472.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bigpatent">BigPatent</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisum">WikiSum</pwcdataset>
-      <video href="2021.acl-long.472.mp4"/>
     </paper>
     <paper id="473">
       <title>Capturing Relations between Scientific Papers: An Abstractive Model for Related Work Section Generation</title>
@@ -7630,9 +7637,9 @@
       <url hash="83eb8570">2021.acl-long.473</url>
       <doi>10.18653/v1/2021.acl-long.473</doi>
       <bibkey>chen-etal-2021-capturing</bibkey>
+      <video href="2021.acl-long.473.mp4"/>
       <pwccode url="https://github.com/iriscxy/relatedworkgeneration" additional="false">iriscxy/relatedworkgeneration</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/s2orc">S2ORC</pwcdataset>
-      <video href="2021.acl-long.473.mp4"/>
     </paper>
     <paper id="474">
       <title>Focus Attention: Promoting Faithfulness and Diversity in Summarization</title>
@@ -7657,9 +7664,9 @@
       <url hash="b0879cbf">2021.acl-long.475</url>
       <doi>10.18653/v1/2021.acl-long.475</doi>
       <bibkey>xu-lapata-2021-generating</bibkey>
+      <video href="2021.acl-long.475.mp4"/>
       <pwccode url="https://github.com/yumoxu/marge" additional="false">yumoxu/marge</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multi-news">Multi-News</pwcdataset>
-      <video href="2021.acl-long.475.mp4"/>
     </paper>
     <paper id="476">
       <title>Robustifying Multi-hop <fixed-case>QA</fixed-case> through Pseudo-Evidentiality Training</title>
@@ -7672,8 +7679,8 @@
       <url hash="7b9bfc86">2021.acl-long.476</url>
       <doi>10.18653/v1/2021.acl-long.476</doi>
       <bibkey>lee-etal-2021-robustifying</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <video href="2021.acl-long.476.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
     </paper>
     <paper id="477">
       <title>x<fixed-case>M</fixed-case>o<fixed-case>C</fixed-case>o: Cross Momentum Contrastive Learning for Open-Domain Question Answering</title>
@@ -7687,11 +7694,11 @@
       <url hash="1d02e683">2021.acl-long.477</url>
       <doi>10.18653/v1/2021.acl-long.477</doi>
       <bibkey>yang-etal-2021-xmoco</bibkey>
+      <video href="2021.acl-long.477.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
-      <video href="2021.acl-long.477.mp4"/>
     </paper>
     <paper id="478">
       <title>Learn to Resolve Conversational Dependency: A Consistency Training Framework for Conversational Question Answering</title>
@@ -7704,11 +7711,11 @@
       <url hash="83c51f5e">2021.acl-long.478</url>
       <doi>10.18653/v1/2021.acl-long.478</doi>
       <bibkey>kim-etal-2021-learn</bibkey>
+      <video href="2021.acl-long.478.mp4"/>
       <pwccode url="https://github.com/dmis-lab/excord" additional="false">dmis-lab/excord</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/canard">CANARD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coqa">CoQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quac">QuAC</pwcdataset>
-      <video href="2021.acl-long.478.mp4"/>
     </paper>
     <paper id="479">
       <title><fixed-case>P</fixed-case>hoto<fixed-case>C</fixed-case>hat: A Human-Human Dialogue Dataset With Photo Sharing Behavior For Joint Image-Text Modeling</title>
@@ -7723,9 +7730,9 @@
       <url hash="7eed00ed">2021.acl-long.479</url>
       <doi>10.18653/v1/2021.acl-long.479</doi>
       <bibkey>zang-etal-2021-photochat</bibkey>
+      <video href="2021.acl-long.479.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.acl-long.479.mp4"/>
     </paper>
     <paper id="480">
       <title>Good for Misconceived Reasons: An Empirical Revisiting on the Need for Visual Context in Multimodal Machine Translation</title>
@@ -7739,8 +7746,8 @@
       <url hash="1858fab0">2021.acl-long.480</url>
       <doi>10.18653/v1/2021.acl-long.480</doi>
       <bibkey>wu-etal-2021-good</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/vatex">VATEX</pwcdataset>
       <video href="2021.acl-long.480.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/vatex">VATEX</pwcdataset>
     </paper>
     <paper id="481">
       <title>Attend What You Need: Motion-Appearance Synergistic Networks for Video Question Answering</title>
@@ -7753,9 +7760,9 @@
       <url hash="948bbd2a">2021.acl-long.481</url>
       <doi>10.18653/v1/2021.acl-long.481</doi>
       <bibkey>seo-etal-2021-attend</bibkey>
+      <video href="2021.acl-long.481.mp4"/>
       <pwccode url="https://github.com/ahjeongseo/MASN-pytorch" additional="false">ahjeongseo/MASN-pytorch</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.acl-long.481.mp4"/>
     </paper>
     <paper id="482">
       <title><fixed-case>BERT</fixed-case>ifying the Hidden <fixed-case>M</fixed-case>arkov Model for Multi-Source Weakly Supervised Named Entity Recognition</title>
@@ -7769,11 +7776,11 @@
       <url hash="6756d039">2021.acl-long.482</url>
       <doi>10.18653/v1/2021.acl-long.482</doi>
       <bibkey>li-etal-2021-bertifying</bibkey>
+      <video href="2021.acl-long.482.mp4"/>
       <pwccode url="https://github.com/Yinghao-Li/CHMM-ALT" additional="false">Yinghao-Li/CHMM-ALT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bc5cdr">BC5CDR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ncbi-disease-1">NCBI Disease</pwcdataset>
-      <video href="2021.acl-long.482.mp4"/>
     </paper>
     <paper id="483">
       <title><fixed-case>CIL</fixed-case>: Contrastive Instance Learning Framework for Distantly Supervised Relation Extraction</title>
@@ -7788,8 +7795,8 @@
       <url hash="ee751e9e">2021.acl-long.483</url>
       <doi>10.18653/v1/2021.acl-long.483</doi>
       <bibkey>chen-etal-2021-cil</bibkey>
-      <pwccode url="https://github.com/antct/cil" additional="false">antct/cil</pwccode>
       <video href="2021.acl-long.483.mp4"/>
+      <pwccode url="https://github.com/antct/cil" additional="false">antct/cil</pwccode>
     </paper>
     <paper id="484">
       <title><fixed-case>SENT</fixed-case>: <fixed-case>S</fixed-case>entence-level Distant Relation Extraction via Negative Training</title>
@@ -7804,8 +7811,8 @@
       <url hash="6c6e6a7c">2021.acl-long.484</url>
       <doi>10.18653/v1/2021.acl-long.484</doi>
       <bibkey>ma-etal-2021-sent</bibkey>
-      <pwccode url="https://github.com/rtmaww/SENT" additional="false">rtmaww/SENT</pwccode>
       <video href="2021.acl-long.484.mp4"/>
+      <pwccode url="https://github.com/rtmaww/SENT" additional="false">rtmaww/SENT</pwccode>
     </paper>
     <paper id="485">
       <title>An End-to-End Progressive Multi-Task Learning Framework for Medical Named Entity Recognition and Normalization</title>
@@ -7818,9 +7825,9 @@
       <url hash="4a2af423">2021.acl-long.485</url>
       <doi>10.18653/v1/2021.acl-long.485</doi>
       <bibkey>zhou-etal-2021-end</bibkey>
+      <video href="2021.acl-long.485.mp4"/>
       <pwccode url="https://github.com/zhoubaohang/e2emern" additional="false">zhoubaohang/e2emern</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bc5cdr">BC5CDR</pwcdataset>
-      <video href="2021.acl-long.485.mp4"/>
     </paper>
     <paper id="486">
       <title><fixed-case>PRGC</fixed-case>: Potential Relation and Global Correspondence Based Joint Relational Triple Extraction</title>
@@ -7840,9 +7847,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="00836461">2021.acl-long.486.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.486</doi>
       <bibkey>zheng-etal-2021-prgc</bibkey>
+      <video href="2021.acl-long.486.mp4"/>
       <pwccode url="https://github.com/hy-struggle/PRGC" additional="false">hy-struggle/PRGC</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.acl-long.486.mp4"/>
     </paper>
     <paper id="487">
       <title>Learning from Miscellaneous Other-Class Words for Few-shot Named Entity Recognition</title>
@@ -7858,9 +7865,9 @@
       <url hash="66f5ff10">2021.acl-long.487</url>
       <doi>10.18653/v1/2021.acl-long.487</doi>
       <bibkey>tong-etal-2021-learning</bibkey>
+      <video href="2021.acl-long.487.mp4"/>
       <pwccode url="https://github.com/shuaiwa16/OtherClassNER" additional="false">shuaiwa16/OtherClassNER</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cluener2020">CLUENER2020</pwcdataset>
-      <video href="2021.acl-long.487.mp4"/>
     </paper>
     <paper id="488">
       <title>Joint Biomedical Entity and Relation Extraction with Knowledge-Enhanced Collective Inference</title>
@@ -7873,8 +7880,8 @@
       <url hash="410cd874">2021.acl-long.488</url>
       <doi>10.18653/v1/2021.acl-long.488</doi>
       <bibkey>lai-etal-2021-joint</bibkey>
-      <pwccode url="https://github.com/laituan245/bio_relex" additional="false">laituan245/bio_relex</pwccode>
       <video href="2021.acl-long.488.mp4"/>
+      <pwccode url="https://github.com/laituan245/bio_relex" additional="false">laituan245/bio_relex</pwccode>
     </paper>
     <paper id="489">
       <title>Fine-grained Information Extraction from Biomedical Literature based on Knowledge-enriched <fixed-case>A</fixed-case>bstract <fixed-case>M</fixed-case>eaning <fixed-case>R</fixed-case>epresentation</title>
@@ -7889,8 +7896,8 @@
       <url hash="0658e0ce">2021.acl-long.489</url>
       <doi>10.18653/v1/2021.acl-long.489</doi>
       <bibkey>zhang-etal-2021-fine</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/bio">Bio</pwcdataset>
       <video href="2021.acl-long.489.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/bio">Bio</pwcdataset>
     </paper>
     <paper id="490">
       <title>Unleash <fixed-case>GPT</fixed-case>-2 Power for Event Detection</title>
@@ -7921,10 +7928,10 @@
       <url hash="40e9237f">2021.acl-long.491</url>
       <doi>10.18653/v1/2021.acl-long.491</doi>
       <bibkey>wang-etal-2021-cleve</bibkey>
+      <video href="2021.acl-long.491.mp4"/>
       <pwccode url="https://github.com/THU-KEG/CLEVE" additional="false">THU-KEG/CLEVE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/maven">MAVEN</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/new-york-times-annotated-corpus">New York Times Annotated Corpus</pwcdataset>
-      <video href="2021.acl-long.491.mp4"/>
     </paper>
     <paper id="492">
       <title>Document-level Event Extraction via Parallel Prediction Networks</title>
@@ -7939,9 +7946,9 @@
       <url hash="cc35c57c">2021.acl-long.492</url>
       <doi>10.18653/v1/2021.acl-long.492</doi>
       <bibkey>yang-etal-2021-document</bibkey>
+      <video href="2021.acl-long.492.mp4"/>
       <pwccode url="https://github.com/hangyang-nlp/de-ppn" additional="true">hangyang-nlp/de-ppn</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/chfinann">ChFinAnn</pwcdataset>
-      <video href="2021.acl-long.492.mp4"/>
     </paper>
     <paper id="493">
       <title><fixed-case>S</fixed-case>tructural<fixed-case>LM</fixed-case>: Structural Pre-training for Form Understanding</title>
@@ -7957,11 +7964,11 @@
       <url hash="01b94570">2021.acl-long.493</url>
       <doi>10.18653/v1/2021.acl-long.493</doi>
       <bibkey>li-etal-2021-structurallm</bibkey>
+      <video href="2021.acl-long.493.mp4"/>
       <pwccode url="https://github.com/alibaba/AliceMind" additional="false">alibaba/AliceMind</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/docvqa">DocVQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/funsd">FUNSD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rvl-cdip">RVL-CDIP</pwcdataset>
-      <video href="2021.acl-long.493.mp4"/>
     </paper>
     <paper id="494">
       <title>Dual Graph Convolutional Networks for Aspect-based Sentiment Analysis</title>
@@ -7976,8 +7983,8 @@
       <url hash="7431232e">2021.acl-long.494</url>
       <doi>10.18653/v1/2021.acl-long.494</doi>
       <bibkey>li-etal-2021-dual-graph</bibkey>
-      <pwccode url="https://github.com/ccchenhao997/dualgcn-absa" additional="false">ccchenhao997/dualgcn-absa</pwccode>
       <video href="2021.acl-long.494.mp4"/>
+      <pwccode url="https://github.com/ccchenhao997/dualgcn-absa" additional="false">ccchenhao997/dualgcn-absa</pwccode>
     </paper>
     <paper id="495">
       <title>Multi-Label Few-Shot Learning for Aspect Category Detection</title>
@@ -8008,9 +8015,9 @@
       <url hash="4976b497">2021.acl-long.496</url>
       <doi>10.18653/v1/2021.acl-long.496</doi>
       <bibkey>cheng-etal-2021-argument</bibkey>
+      <video href="2021.acl-long.496.mp4"/>
       <pwccode url="https://github.com/tianyuterry/mlmc" additional="false">tianyuterry/mlmc</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/rr">RR</pwcdataset>
-      <video href="2021.acl-long.496.mp4"/>
     </paper>
     <paper id="497">
       <title>A Neural Transition-based Model for Argumentation Mining</title>
@@ -8025,8 +8032,8 @@
       <url hash="ff56cd95">2021.acl-long.497</url>
       <doi>10.18653/v1/2021.acl-long.497</doi>
       <bibkey>bao-etal-2021-neural</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/cdcp">CDCP</pwcdataset>
       <video href="2021.acl-long.497.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cdcp">CDCP</pwcdataset>
     </paper>
     <paper id="498">
       <title>Keep It Simple: Unsupervised Simplification of Multi-Paragraph Text</title>
@@ -8039,10 +8046,10 @@
       <url hash="ef17e0a9">2021.acl-long.498</url>
       <doi>10.18653/v1/2021.acl-long.498</doi>
       <bibkey>laban-etal-2021-keep</bibkey>
+      <video href="2021.acl-long.498.mp4"/>
       <pwccode url="https://github.com/tingofurro/keep_it_simple" additional="false">tingofurro/keep_it_simple</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/newsela">Newsela</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikilarge">WikiLarge</pwcdataset>
-      <video href="2021.acl-long.498.mp4"/>
     </paper>
     <paper id="499">
       <title>Long Text Generation by Modeling Sentence-Level and Discourse-Level Coherence</title>
@@ -8057,11 +8064,11 @@
       <url hash="107cae2f">2021.acl-long.499</url>
       <doi>10.18653/v1/2021.acl-long.499</doi>
       <bibkey>guan-etal-2021-long</bibkey>
+      <video href="2021.acl-long.499.mp4"/>
       <pwccode url="https://github.com/thu-coai/HINT" additional="false">thu-coai/HINT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
-      <video href="2021.acl-long.499.mp4"/>
     </paper>
     <paper id="500">
       <title><fixed-case>O</fixed-case>pen<fixed-case>MEVA</fixed-case>: A Benchmark for Evaluating Open-ended Story Generation Metrics</title>
@@ -8078,11 +8085,11 @@
       <url hash="446f1817">2021.acl-long.500</url>
       <doi>10.18653/v1/2021.acl-long.500</doi>
       <bibkey>guan-etal-2021-openmeva</bibkey>
+      <video href="2021.acl-long.500.mp4"/>
       <pwccode url="https://github.com/thu-coai/OpenMEVA" additional="false">thu-coai/OpenMEVA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/openmeva">OpenMEVA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
-      <video href="2021.acl-long.500.mp4"/>
     </paper>
     <paper id="501">
       <title><fixed-case>DYPLOC</fixed-case>: Dynamic Planning of Content Using Mixed Language Models for Text Generation</title>
@@ -8094,8 +8101,8 @@
       <url hash="922b8d58">2021.acl-long.501</url>
       <doi>10.18653/v1/2021.acl-long.501</doi>
       <bibkey>hua-etal-2021-dyploc</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <video href="2021.acl-long.501.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
     </paper>
     <paper id="502">
       <title>Controllable Open-ended Question Generation with A New Question Type Ontology</title>
@@ -8106,10 +8113,10 @@
       <url hash="c3e7fb0d">2021.acl-long.502</url>
       <doi>10.18653/v1/2021.acl-long.502</doi>
       <bibkey>cao-wang-2021-controllable</bibkey>
+      <video href="2021.acl-long.502.mp4"/>
       <pwccode url="https://github.com/ShuyangCao/open-ended_question_ontology" additional="false">ShuyangCao/open-ended_question_ontology</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.502.mp4"/>
     </paper>
     <paper id="503">
       <title><fixed-case>BERTG</fixed-case>en: Multi-task Generation through <fixed-case>BERT</fixed-case></title>
@@ -8122,8 +8129,8 @@
       <url hash="e22305bd">2021.acl-long.503</url>
       <doi>10.18653/v1/2021.acl-long.503</doi>
       <bibkey>mitzalis-etal-2021-bertgen</bibkey>
-      <pwccode url="https://github.com/ImperialNLP/BertGen" additional="false">ImperialNLP/BertGen</pwccode>
       <video href="2021.acl-long.503.mp4"/>
+      <pwccode url="https://github.com/ImperialNLP/BertGen" additional="false">ImperialNLP/BertGen</pwccode>
     </paper>
     <paper id="504">
       <title>Selective Knowledge Distillation for Neural Machine Translation</title>
@@ -8136,8 +8143,8 @@
       <url hash="44c4c211">2021.acl-long.504</url>
       <doi>10.18653/v1/2021.acl-long.504</doi>
       <bibkey>wang-etal-2021-selective</bibkey>
-      <pwccode url="https://github.com/LeslieOverfitting/selective_distillation" additional="false">LeslieOverfitting/selective_distillation</pwccode>
       <video href="2021.acl-long.504.mp4"/>
+      <pwccode url="https://github.com/LeslieOverfitting/selective_distillation" additional="false">LeslieOverfitting/selective_distillation</pwccode>
     </paper>
     <paper id="505">
       <title>Measuring and Increasing Context Usage in Context-Aware Machine Translation</title>
@@ -8150,8 +8157,8 @@
       <url hash="d2043bd4">2021.acl-long.505</url>
       <doi>10.18653/v1/2021.acl-long.505</doi>
       <bibkey>fernandes-etal-2021-measuring</bibkey>
-      <pwccode url="https://github.com/neulab/contextual-mt" additional="false">neulab/contextual-mt</pwccode>
       <video href="2021.acl-long.505.mp4"/>
+      <pwccode url="https://github.com/neulab/contextual-mt" additional="false">neulab/contextual-mt</pwccode>
     </paper>
     <paper id="506">
       <title>Beyond Offline Mapping: Learning Cross-lingual Word Embeddings through Context Anchoring</title>
@@ -8165,8 +8172,8 @@
       <url hash="a04ea4ea">2021.acl-long.506</url>
       <doi>10.18653/v1/2021.acl-long.506</doi>
       <bibkey>ormazabal-etal-2021-beyond</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <video href="2021.acl-long.506.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
     </paper>
     <paper id="507">
       <title><fixed-case>CCM</fixed-case>atrix: Mining Billions of High-Quality Parallel Sentences on the Web</title>
@@ -8182,10 +8189,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="86df1c16">2021.acl-long.507.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.507</doi>
       <bibkey>schwenk-etal-2021-ccmatrix</bibkey>
+      <video href="2021.acl-long.507.mp4"/>
       <pwccode url="https://github.com/facebookresearch/LASER" additional="true">facebookresearch/LASER</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ccmatrix">CCMatrix</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ccnet">CCNet</pwcdataset>
-      <video href="2021.acl-long.507.mp4"/>
     </paper>
     <paper id="508">
       <title>Length-Adaptive Transformer: Train Once with Length Drop, Use Anytime with Search</title>
@@ -8196,10 +8203,10 @@
       <url hash="70cdff0a">2021.acl-long.508</url>
       <doi>10.18653/v1/2021.acl-long.508</doi>
       <bibkey>kim-cho-2021-length</bibkey>
+      <video href="2021.acl-long.508.mp4"/>
       <pwccode url="https://github.com/clovaai/length-adaptive-transformer" additional="false">clovaai/length-adaptive-transformer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.508.mp4"/>
     </paper>
     <paper id="509">
       <title><fixed-case>G</fixed-case>host<fixed-case>BERT</fixed-case>: Generate More Features with Cheap Operations for <fixed-case>BERT</fixed-case></title>
@@ -8214,9 +8221,9 @@
       <url hash="64fccf67">2021.acl-long.509</url>
       <doi>10.18653/v1/2021.acl-long.509</doi>
       <bibkey>huang-etal-2021-ghostbert</bibkey>
+      <video href="2021.acl-long.509.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-long.509.mp4"/>
     </paper>
     <paper id="510">
       <title>Super Tickets in Pre-Trained Language Models: From Model Compression to Improving Generalization</title>
@@ -8233,6 +8240,7 @@
       <url hash="0f7a8ee7">2021.acl-long.510</url>
       <doi>10.18653/v1/2021.acl-long.510</doi>
       <bibkey>liang-etal-2021-super</bibkey>
+      <video href="2021.acl-long.510.mp4"/>
       <pwccode url="https://github.com/cliang1453/super-structured-lottery-tickets" additional="false">cliang1453/super-structured-lottery-tickets</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
@@ -8240,7 +8248,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.510.mp4"/>
     </paper>
     <paper id="511">
       <title>A Novel Estimator of Mutual Information for Learning to Disentangle Textual Representations</title>
@@ -8265,8 +8272,8 @@
       <url hash="f7dfc94b">2021.acl-long.512</url>
       <doi>10.18653/v1/2021.acl-long.512</doi>
       <bibkey>meister-etal-2021-determinantal</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
       <video href="2021.acl-long.512.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
     </paper>
     <paper id="513">
       <title>Multi-hop Graph Convolutional Network with High-order Chebyshev Approximation for Text Reasoning</title>
@@ -8280,9 +8287,9 @@
       <url hash="1d7b0ec7">2021.acl-long.513</url>
       <doi>10.18653/v1/2021.acl-long.513</doi>
       <bibkey>jiang-etal-2021-multi</bibkey>
+      <video href="2021.acl-long.513.mp4"/>
       <pwccode url="https://github.com/MathIsAll/HDGCN-pytorch" additional="false">MathIsAll/HDGCN-pytorch</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.acl-long.513.mp4"/>
     </paper>
     <paper id="514">
       <title>Accelerating Text Communication via Abbreviated Sentence Input</title>
@@ -8295,9 +8302,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="0edd8db0">2021.acl-long.514.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.514</doi>
       <bibkey>adhikary-etal-2021-accelerating</bibkey>
+      <video href="2021.acl-long.514.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.acl-long.514.mp4"/>
     </paper>
     <paper id="515">
       <title>Regression Bugs Are In Your Model! Measuring, Reducing and Analyzing Regressions In <fixed-case>NLP</fixed-case> Model Updates</title>
@@ -8311,11 +8318,11 @@
       <url hash="157e438a">2021.acl-long.515</url>
       <doi>10.18653/v1/2021.acl-long.515</doi>
       <bibkey>xie-etal-2021-regression</bibkey>
+      <video href="2021.acl-long.515.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.515.mp4"/>
     </paper>
     <paper id="516">
       <title>Detecting Propaganda Techniques in Memes</title>
@@ -8332,10 +8339,10 @@
       <url hash="007ff28a">2021.acl-long.516</url>
       <doi>10.18653/v1/2021.acl-long.516</doi>
       <bibkey>dimitrov-etal-2021-detecting</bibkey>
+      <video href="2021.acl-long.516.mp4"/>
       <pwccode url="https://github.com/di-dimitrov/propaganda-techniques-in-memes" additional="false">di-dimitrov/propaganda-techniques-in-memes</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hateful-memes">Hateful Memes</pwcdataset>
-      <video href="2021.acl-long.516.mp4"/>
     </paper>
     <paper id="517">
       <title>On the Efficacy of Adversarial Data Collection for Question Answering: Results from a Large-Scale Randomized Study</title>
@@ -8348,6 +8355,7 @@
       <url hash="26907685">2021.acl-long.517</url>
       <doi>10.18653/v1/2021.acl-long.517</doi>
       <bibkey>kaushik-etal-2021-efficacy</bibkey>
+      <video href="2021.acl-long.517.mp4"/>
       <pwccode url="https://github.com/facebookresearch/aqa-study" additional="false">facebookresearch/aqa-study</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
@@ -8358,7 +8366,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/searchqa">SearchQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.acl-long.517.mp4"/>
     </paper>
     <paper id="518">
       <title>Learning Dense Representations of Phrases at Scale</title>
@@ -8371,13 +8378,13 @@
       <url hash="59fa12e7">2021.acl-long.518</url>
       <doi>10.18653/v1/2021.acl-long.518</doi>
       <bibkey>lee-etal-2021-learning-dense</bibkey>
+      <video href="2021.acl-long.518.mp4"/>
       <pwccode url="https://github.com/jhyuklee/DensePhrases" additional="true">jhyuklee/DensePhrases</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
-      <video href="2021.acl-long.518.mp4"/>
     </paper>
     <paper id="519">
       <title>End-to-End Training of Neural Retrievers for Open-Domain Question Answering</title>
@@ -8393,10 +8400,10 @@
       <url hash="c27e7f8c">2021.acl-long.519</url>
       <doi>10.18653/v1/2021.acl-long.519</doi>
       <bibkey>sachan-etal-2021-end</bibkey>
+      <video href="2021.acl-long.519.mp4"/>
       <pwccode url="https://github.com/NVIDIA/Megatron-LM" additional="false">NVIDIA/Megatron-LM</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.acl-long.519.mp4"/>
     </paper>
     <paper id="520">
       <title>Question Answering Over Temporal Knowledge Graphs</title>
@@ -8408,12 +8415,12 @@
       <url hash="672feef4">2021.acl-long.520</url>
       <doi>10.18653/v1/2021.acl-long.520</doi>
       <bibkey>saxena-etal-2021-question</bibkey>
+      <video href="2021.acl-long.520.mp4"/>
       <pwccode url="https://github.com/apoorvumang/CronKGQA" additional="true">apoorvumang/CronKGQA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cronquestions">CronQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/complexwebquestions">ComplexWebQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/metaqa">MetaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/simplequestions">SimpleQuestions</pwcdataset>
-      <video href="2021.acl-long.520.mp4"/>
     </paper>
     <paper id="521">
       <title>Language Model Augmented Relevance Score</title>
@@ -8425,9 +8432,9 @@
       <url hash="88b302d1">2021.acl-long.521</url>
       <doi>10.18653/v1/2021.acl-long.521</doi>
       <bibkey>liu-etal-2021-language</bibkey>
+      <video href="2021.acl-long.521.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/mocha">MOCHA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsroom">NEWSROOM</pwcdataset>
-      <video href="2021.acl-long.521.mp4"/>
     </paper>
     <paper id="522">
       <title><fixed-case>DE</fixed-case>xperts: Decoding-Time Controlled Text Generation with Experts and Anti-Experts</title>
@@ -8443,11 +8450,11 @@
       <url hash="b6cc3492">2021.acl-long.522</url>
       <doi>10.18653/v1/2021.acl-long.522</doi>
       <bibkey>liu-etal-2021-dexperts</bibkey>
+      <video href="2021.acl-long.522.mp4"/>
       <pwccode url="https://github.com/alisawuffles/DExperts" additional="true">alisawuffles/DExperts</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/openwebtext">OpenWebText</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.acl-long.522.mp4"/>
     </paper>
     <paper id="523">
       <title>Polyjuice: Generating Counterfactuals for Explaining, Evaluating, and Improving Models</title>
@@ -8460,11 +8467,11 @@
       <url hash="5846f631">2021.acl-long.523</url>
       <doi>10.18653/v1/2021.acl-long.523</doi>
       <bibkey>wu-etal-2021-polyjuice</bibkey>
+      <video href="2021.acl-long.523.mp4"/>
       <pwccode url="https://github.com/tongshuangwu/polyjuice" additional="false">tongshuangwu/polyjuice</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-long.523.mp4"/>
     </paper>
     <paper id="524">
       <title>Metaphor Generation with Conceptual Mappings</title>
@@ -8478,9 +8485,9 @@
       <url hash="580635a6">2021.acl-long.524</url>
       <doi>10.18653/v1/2021.acl-long.524</doi>
       <bibkey>stowe-etal-2021-metaphor</bibkey>
+      <video href="2021.acl-long.524.mp4"/>
       <pwccode url="https://github.com/UKPLab/acl2021-metaphor-generation-conceptual" additional="false">UKPLab/acl2021-metaphor-generation-conceptual</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
-      <video href="2021.acl-long.524.mp4"/>
     </paper>
     <paper id="525">
       <title>Learning Latent Structures for Cross Action Phrase Relations in Wet Lab Protocols</title>
@@ -8494,8 +8501,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="4e61f091">2021.acl-long.525.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.525</doi>
       <bibkey>kulkarni-etal-2021-learning</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wnut-20-task-1-extracting-entities-and">WNUT 2020</pwcdataset>
       <video href="2021.acl-long.525.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wnut-20-task-1-extracting-entities-and">WNUT 2020</pwcdataset>
     </paper>
     <paper id="526">
       <title>Multimodal Multi-Speaker Merger &amp; Acquisition Financial Modeling: A New Task, Dataset, and Neural Baselines</title>
@@ -8522,8 +8529,8 @@
       <url hash="eba54d17">2021.acl-long.527</url>
       <doi>10.18653/v1/2021.acl-long.527</doi>
       <bibkey>albo-jamara-etal-2021-mid</bibkey>
-      <pwccode url="https://github.com/nicoherbig/mmpe" additional="false">nicoherbig/mmpe</pwccode>
       <video href="2021.acl-long.527.mp4"/>
+      <pwccode url="https://github.com/nicoherbig/mmpe" additional="false">nicoherbig/mmpe</pwccode>
     </paper>
     <paper id="528">
       <title><fixed-case>I</fixed-case>nter-<fixed-case>GPS</fixed-case>: Interpretable Geometry Problem Solving with Formal Language and Symbolic Reasoning</title>
@@ -8539,11 +8546,11 @@
       <url hash="f62be951">2021.acl-long.528</url>
       <doi>10.18653/v1/2021.acl-long.528</doi>
       <bibkey>lu-etal-2021-inter</bibkey>
+      <video href="2021.acl-long.528.mp4"/>
       <pwccode url="https://github.com/lupantech/InterGPS" additional="false">lupantech/InterGPS</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/geometry3k">Geometry3K</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/geos">GeoS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/pgdp5k">PGDP5K</pwcdataset>
-      <video href="2021.acl-long.528.mp4"/>
     </paper>
     <paper id="529">
       <title>Joint Verification and Reranking for Open Fact Checking Over Tables</title>
@@ -8558,8 +8565,8 @@
       <url hash="93c56769">2021.acl-long.529</url>
       <doi>10.18653/v1/2021.acl-long.529</doi>
       <bibkey>schlichtkrull-etal-2021-joint</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/tabfact">TabFact</pwcdataset>
       <video href="2021.acl-long.529.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/tabfact">TabFact</pwcdataset>
     </paper>
     <paper id="530">
       <title>Evaluation of Thematic Coherence in Microblogs</title>
@@ -8573,8 +8580,8 @@
       <url hash="ba881b2e">2021.acl-long.530</url>
       <doi>10.18653/v1/2021.acl-long.530</doi>
       <bibkey>bilal-etal-2021-evaluation</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/covid-19-election">COVID-19 &amp; Election</pwcdataset>
       <video href="2021.acl-long.530.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/covid-19-election">COVID-19 &amp; Election</pwcdataset>
     </paper>
     <paper id="531">
       <title>Neural semi-<fixed-case>M</fixed-case>arkov <fixed-case>CRF</fixed-case> for Monolingual Word Alignment</title>
@@ -8586,6 +8593,7 @@
       <url hash="a74fc606">2021.acl-long.531</url>
       <doi>10.18653/v1/2021.acl-long.531</doi>
       <bibkey>lan-etal-2021-neural</bibkey>
+      <video href="2021.acl-long.531.mp4"/>
       <pwccode url="https://github.com/chaojiang06/neural-Jacana" additional="false">chaojiang06/neural-Jacana</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
@@ -8594,7 +8602,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/sick">SICK</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikiqa">WikiQA</pwcdataset>
-      <video href="2021.acl-long.531.mp4"/>
     </paper>
     <paper id="532">
       <title>Privacy at Scale: Introducing the <fixed-case>P</fixed-case>riva<fixed-case>S</fixed-case>eer Corpus of Web Privacy Policies</title>
@@ -8617,8 +8624,8 @@
       <url hash="5be2d1f9">2021.acl-long.533</url>
       <doi>10.18653/v1/2021.acl-long.533</doi>
       <bibkey>wei-jia-2021-statistical</bibkey>
-      <pwccode url="https://github.com/johntzwei/metric-statistical-advantage" additional="false">johntzwei/metric-statistical-advantage</pwccode>
       <video href="2021.acl-long.533.mp4"/>
+      <pwccode url="https://github.com/johntzwei/metric-statistical-advantage" additional="false">johntzwei/metric-statistical-advantage</pwccode>
     </paper>
     <paper id="534">
       <title>Are Missing Links Predictable? An Inferential Benchmark for Knowledge Graph Completion</title>
@@ -8634,10 +8641,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="81a43de2">2021.acl-long.534.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-long.534</doi>
       <bibkey>cao-etal-2021-missing</bibkey>
+      <video href="2021.acl-long.534.mp4"/>
       <pwccode url="https://github.com/TaoMiner/inferwiki" additional="false">TaoMiner/inferwiki</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/inferwiki">InferWiki</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
-      <video href="2021.acl-long.534.mp4"/>
     </paper>
     <paper id="535">
       <title><fixed-case>C</fixed-case>onvo<fixed-case>S</fixed-case>umm: Conversation Summarization Benchmark and Improved Abstractive Summarization with Argument Mining</title>
@@ -8653,12 +8660,12 @@
       <url hash="3a8f0363">2021.acl-long.535</url>
       <doi>10.18653/v1/2021.acl-long.535</doi>
       <bibkey>fabbri-etal-2021-convosumm</bibkey>
+      <video href="2021.acl-long.535.mp4"/>
       <pwccode url="https://github.com/Yale-LILY/ConvoSumm" additional="false">Yale-LILY/ConvoSumm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/convosumm">ConvoSumm</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cqasumm">CQASUMM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.acl-long.535.mp4"/>
     </paper>
     <paper id="536">
       <title>Improving Factual Consistency of Abstractive Summarization via Question Answering</title>
@@ -8677,8 +8684,8 @@
       <url hash="5e076d13">2021.acl-long.536</url>
       <doi>10.18653/v1/2021.acl-long.536</doi>
       <bibkey>nan-etal-2021-improving</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <video href="2021.acl-long.536.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
     </paper>
     <paper id="537">
       <title><fixed-case>E</fixed-case>mail<fixed-case>S</fixed-case>um: Abstractive Email Thread Summarization</title>
@@ -8691,11 +8698,11 @@
       <url hash="e674332b">2021.acl-long.537</url>
       <doi>10.18653/v1/2021.acl-long.537</doi>
       <bibkey>zhang-etal-2021-emailsum</bibkey>
+      <video href="2021.acl-long.537.mp4"/>
       <pwccode url="https://github.com/ZhangShiyue/EmailSum" additional="false">ZhangShiyue/EmailSum</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/emailsum">EmailSum</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/crd3">CRD3</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.acl-long.537.mp4"/>
     </paper>
     <paper id="538">
       <title>Cross-Lingual Abstractive Summarization with Limited Parallel Resources</title>
@@ -8707,9 +8714,9 @@
       <url hash="63d6fe54">2021.acl-long.538</url>
       <doi>10.18653/v1/2021.acl-long.538</doi>
       <bibkey>bai-etal-2021-cross</bibkey>
+      <video href="2021.acl-long.538.mp4"/>
       <pwccode url="https://github.com/WoodenWhite/MCLAS" additional="false">WoodenWhite/MCLAS</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ncls">NCLS</pwcdataset>
-      <video href="2021.acl-long.538.mp4"/>
     </paper>
     <paper id="539">
       <title>Dissecting Generation Modes for Abstractive Summarization Models via Ablation and Attribution</title>
@@ -8720,10 +8727,10 @@
       <url hash="6b15ade6">2021.acl-long.539</url>
       <doi>10.18653/v1/2021.acl-long.539</doi>
       <bibkey>xu-durrett-2021-dissecting</bibkey>
+      <video href="2021.acl-long.539.mp4"/>
       <pwccode url="https://github.com/jiacheng-xu/sum-interpret" additional="false">jiacheng-xu/sum-interpret</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.acl-long.539.mp4"/>
     </paper>
     <paper id="540">
       <title>Learning Prototypical Functions for Physical Artifacts</title>
@@ -8734,10 +8741,10 @@
       <url hash="932892c3">2021.acl-long.540</url>
       <doi>10.18653/v1/2021.acl-long.540</doi>
       <bibkey>jiang-riloff-2021-learning</bibkey>
+      <video href="2021.acl-long.540.mp4"/>
       <pwccode url="https://github.com/tyjiangu/physical_artifacts_function" additional="false">tyjiangu/physical_artifacts_function</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
-      <video href="2021.acl-long.540.mp4"/>
     </paper>
     <paper id="541">
       <title>Verb Knowledge Injection for Multilingual Event Processing</title>
@@ -8751,9 +8758,9 @@
       <url hash="d0f6ab90">2021.acl-long.541</url>
       <doi>10.18653/v1/2021.acl-long.541</doi>
       <bibkey>majewska-etal-2021-verb</bibkey>
+      <video href="2021.acl-long.541.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/maven">MAVEN</pwcdataset>
-      <video href="2021.acl-long.541.mp4"/>
     </paper>
     <paper id="542">
       <title>Dynamic Contextualized Word Embeddings</title>
@@ -8765,9 +8772,9 @@
       <url hash="080ba29b">2021.acl-long.542</url>
       <doi>10.18653/v1/2021.acl-long.542</doi>
       <bibkey>hofmann-etal-2021-dynamic</bibkey>
+      <video href="2021.acl-long.542.mp4"/>
       <pwccode url="https://github.com/valentinhofmann/dcwe" additional="false">valentinhofmann/dcwe</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ciao">Ciao</pwcdataset>
-      <video href="2021.acl-long.542.mp4"/>
     </paper>
     <paper id="543">
       <title>Lexical Semantic Change Discovery</title>
@@ -8781,8 +8788,8 @@
       <url hash="b08914c5">2021.acl-long.543</url>
       <doi>10.18653/v1/2021.acl-long.543</doi>
       <bibkey>kurtyigit-etal-2021-lexical</bibkey>
-      <pwccode url="https://github.com/seinan9/LSCDiscovery" additional="false">seinan9/LSCDiscovery</pwccode>
       <video href="2021.acl-long.543.mp4"/>
+      <pwccode url="https://github.com/seinan9/LSCDiscovery" additional="false">seinan9/LSCDiscovery</pwccode>
     </paper>
     <paper id="544">
       <title>The <fixed-case>R</fixed-case>-<fixed-case>U</fixed-case>-A-Robot Dataset: Helping Avoid Chatbot Deception by Detecting User Questions About Human or Non-Human Identity</title>
@@ -8794,9 +8801,9 @@
       <url hash="d893de79">2021.acl-long.544</url>
       <doi>10.18653/v1/2021.acl-long.544</doi>
       <bibkey>gros-etal-2021-r</bibkey>
+      <video href="2021.acl-long.544.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ethics-1">ETHICS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.acl-long.544.mp4"/>
     </paper>
     <paper id="545">
       <title>Using Meta-Knowledge Mined from Identifiers to Improve Intent Recognition in Conversational Systems</title>
@@ -8829,9 +8836,9 @@
       <url hash="d71dc0fc">2021.acl-long.546</url>
       <doi>10.18653/v1/2021.acl-long.546</doi>
       <bibkey>galetzka-etal-2021-space</bibkey>
+      <video href="2021.acl-long.546.mp4"/>
       <pwccode url="https://github.com/fabiangal/space-efficient-context-encoding-acl21" additional="false">fabiangal/space-efficient-context-encoding-acl21</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/opendialkg">OpenDialKG</pwcdataset>
-      <video href="2021.acl-long.546.mp4"/>
     </paper>
     <paper id="547">
       <title><fixed-case>D</fixed-case>ialogue<fixed-case>CRN</fixed-case>: Contextual Reasoning Networks for Emotion Recognition in Conversations</title>
@@ -8843,12 +8850,12 @@
       <url hash="02ed7a08">2021.acl-long.547</url>
       <doi>10.18653/v1/2021.acl-long.547</doi>
       <bibkey>hu-etal-2021-dialoguecrn</bibkey>
+      <video href="2021.acl-long.547.mp4"/>
       <pwccode url="https://github.com/zerohd4869/DialogueCRN" additional="false">zerohd4869/DialogueCRN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/emotionlines">EmotionLines</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semaine">SEMAINE</pwcdataset>
-      <video href="2021.acl-long.547.mp4"/>
     </paper>
     <paper id="548">
       <title>Cross-replication Reliability - An Empirical Approach to Interpreting Inter-rater Reliability</title>
@@ -8860,8 +8867,8 @@
       <url hash="5b939413">2021.acl-long.548</url>
       <doi>10.18653/v1/2021.acl-long.548</doi>
       <bibkey>wong-etal-2021-cross</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/imagenet">ImageNet</pwcdataset>
       <video href="2021.acl-long.548.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/imagenet">ImageNet</pwcdataset>
     </paper>
     <paper id="549">
       <title><fixed-case>TIMEDIAL</fixed-case>: Temporal Commonsense Reasoning in Dialog</title>
@@ -8877,10 +8884,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="a86161f4">2021.acl-long.549.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-long.549</doi>
       <bibkey>qin-etal-2021-timedial</bibkey>
+      <video href="2021.acl-long.549.mp4"/>
       <pwccode url="https://github.com/google-research-datasets/timedial" additional="false">google-research-datasets/timedial</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/timedial">TimeDial</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
-      <video href="2021.acl-long.549.mp4"/>
     </paper>
     <paper id="550">
       <title><fixed-case>RAW</fixed-case>-<fixed-case>C</fixed-case>: Relatedness of Ambiguous Words in Context (A New Lexical Resource for <fixed-case>E</fixed-case>nglish)</title>
@@ -8891,8 +8898,8 @@
       <url hash="da8d7d94">2021.acl-long.550</url>
       <doi>10.18653/v1/2021.acl-long.550</doi>
       <bibkey>trott-bergen-2021-raw</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/raw-c">RAW-C</pwcdataset>
       <video href="2021.acl-long.550.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/raw-c">RAW-C</pwcdataset>
     </paper>
     <paper id="551">
       <title><fixed-case>ARBERT</fixed-case> &amp; <fixed-case>MARBERT</fixed-case>: Deep Bidirectional Transformers for <fixed-case>A</fixed-case>rabic</title>
@@ -8904,13 +8911,13 @@
       <url hash="789b4ded">2021.acl-long.551</url>
       <doi>10.18653/v1/2021.acl-long.551</doi>
       <bibkey>abdul-mageed-etal-2021-arbert</bibkey>
+      <video href="2021.acl-long.551.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/arcd">ARCD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/astd">ASTD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/labr">LABR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/decanlp">decaNLP</pwcdataset>
-      <video href="2021.acl-long.551.mp4"/>
     </paper>
     <paper id="552">
       <title>Improving Paraphrase Detection with the Adversarial Paraphrasing Task</title>
@@ -8921,6 +8928,7 @@
       <url hash="76048719">2021.acl-long.552</url>
       <doi>10.18653/v1/2021.acl-long.552</doi>
       <bibkey>nighojkar-licato-2021-improving</bibkey>
+      <video href="2021.acl-long.552.mp4"/>
       <pwccode url="https://github.com/Advancing-Machine-Human-Reasoning-Lab/apt" additional="false">Advancing-Machine-Human-Reasoning-Lab/apt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ap">AP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
@@ -8928,7 +8936,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/paws">PAWS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.acl-long.552.mp4"/>
     </paper>
     <paper id="553">
       <title><fixed-case>ADEPT</fixed-case>: An Adjective-Dependent Plausibility Task</title>
@@ -8943,9 +8950,9 @@
       <url hash="5c0773eb">2021.acl-long.553</url>
       <doi>10.18653/v1/2021.acl-long.553</doi>
       <bibkey>emami-etal-2021-adept</bibkey>
+      <video href="2021.acl-long.553.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.acl-long.553.mp4"/>
     </paper>
     <paper id="554">
       <title><fixed-case>R</fixed-case>ead<fixed-case>O</fixed-case>nce Transformers: Reusable Representations of Text for Transformers</title>
@@ -8957,10 +8964,10 @@
       <url hash="76dc9d00">2021.acl-long.554</url>
       <doi>10.18653/v1/2021.acl-long.554</doi>
       <bibkey>lin-etal-2021-readonce</bibkey>
+      <video href="2021.acl-long.554.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/narrativeqa">NarrativeQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-long.554.mp4"/>
     </paper>
     <paper id="555">
       <title>Conditional Generation of Temporally-ordered Event Sequences</title>
@@ -8972,8 +8979,8 @@
       <url hash="309452f6">2021.acl-long.555</url>
       <doi>10.18653/v1/2021.acl-long.555</doi>
       <bibkey>lin-etal-2021-conditional</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mc-taco">MC-TACO</pwcdataset>
       <video href="2021.acl-long.555.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mc-taco">MC-TACO</pwcdataset>
     </paper>
     <paper id="556">
       <title>Hate Speech Detection Based on Sentiment Knowledge Sharing</title>
@@ -8990,8 +8997,8 @@
       <url hash="f31893f2">2021.acl-long.556</url>
       <doi>10.18653/v1/2021.acl-long.556</doi>
       <bibkey>zhou-etal-2021-hate</bibkey>
-      <pwccode url="https://github.com/1783696285/sks" additional="false">1783696285/sks</pwccode>
       <video href="2021.acl-long.556.mp4"/>
+      <pwccode url="https://github.com/1783696285/sks" additional="false">1783696285/sks</pwccode>
     </paper>
     <paper id="557">
       <title>Transition-based Bubble Parsing: Improvements on Coordination Structure Prediction</title>
@@ -9002,11 +9009,11 @@
       <url hash="8d6dfa7c">2021.acl-long.557</url>
       <doi>10.18653/v1/2021.acl-long.557</doi>
       <bibkey>shi-lee-2021-transition</bibkey>
+      <video href="2021.acl-long.557.mp4"/>
       <pwccode url="https://github.com/tzshi/bubble-parser-acl21" additional="false">tzshi/bubble-parser-acl21</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/genia">GENIA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.acl-long.557.mp4"/>
     </paper>
     <paper id="558">
       <title><fixed-case>S</fixed-case>pan<fixed-case>NER</fixed-case>: Named Entity Re-/Recognition as Span Prediction</title>
@@ -9018,11 +9025,11 @@
       <url hash="140c373e">2021.acl-long.558</url>
       <doi>10.18653/v1/2021.acl-long.558</doi>
       <bibkey>fu-etal-2021-spanner</bibkey>
+      <video href="2021.acl-long.558.mp4"/>
       <pwccode url="https://github.com/neulab/spanner" additional="false">neulab/spanner</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2002">CoNLL 2002</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wnut-2016-ner">WNUT 2016 NER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wnut-2017-emerging-and-rare-entity">WNUT 2017</pwcdataset>
-      <video href="2021.acl-long.558.mp4"/>
     </paper>
     <paper id="559">
       <title><fixed-case>S</fixed-case>truct<fixed-case>F</fixed-case>ormer: Joint Unsupervised Induction of Dependency and Constituency Structure from Masked Language Modeling</title>
@@ -9037,8 +9044,8 @@
       <url hash="2533a137">2021.acl-long.559</url>
       <doi>10.18653/v1/2021.acl-long.559</doi>
       <bibkey>shen-etal-2021-structformer</bibkey>
-      <pwccode url="https://github.com/google-research/google-research/tree/master/structformer" additional="false">google-research/google-research</pwccode>
       <video href="2021.acl-long.559.mp4"/>
+      <pwccode url="https://github.com/google-research/google-research/tree/master/structformer" additional="false">google-research/google-research</pwccode>
     </paper>
     <paper id="560">
       <title>Language Embeddings for Typology and Cross-lingual Transfer Learning</title>
@@ -9050,9 +9057,9 @@
       <url hash="e5afacf0">2021.acl-long.560</url>
       <doi>10.18653/v1/2021.acl-long.560</doi>
       <bibkey>yu-etal-2021-language</bibkey>
+      <video href="2021.acl-long.560.mp4"/>
       <pwccode url="https://github.com/DianDYu/language_embeddings" additional="false">DianDYu/language_embeddings</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.acl-long.560.mp4"/>
     </paper>
     <paper id="561">
       <title>Can Sequence-to-Sequence Models Crack Substitution Ciphers?</title>
@@ -9074,9 +9081,9 @@
       <url hash="57d816d3">2021.acl-long.562</url>
       <doi>10.18653/v1/2021.acl-long.562</doi>
       <bibkey>briakou-carpuat-2021-beyond</bibkey>
+      <video href="2021.acl-long.562.mp4"/>
       <pwccode url="https://github.com/awslabs/sockeye" additional="true">awslabs/sockeye</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikimatrix">WikiMatrix</pwcdataset>
-      <video href="2021.acl-long.562.mp4"/>
     </paper>
     <paper id="563">
       <title>Discriminative Reranking for Neural Machine Translation</title>
@@ -9102,11 +9109,11 @@
       <award>Outstanding Paper</award>
       <doi>10.18653/v1/2021.acl-long.564</doi>
       <bibkey>karamcheti-etal-2021-mind</bibkey>
+      <video href="2021.acl-long.564.mp4"/>
       <pwccode url="https://github.com/siddk/vqa-outliers" additional="false">siddk/vqa-outliers</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gqa">GQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.acl-long.564.mp4"/>
     </paper>
     <paper id="565">
       <title>All That’s ‘Human’ Is Not Gold: Evaluating Human Evaluation of Generated Text</title>
@@ -9137,8 +9144,8 @@
       <award>Outstanding Paper</award>
       <doi>10.18653/v1/2021.acl-long.566</doi>
       <bibkey>marie-etal-2021-scientific</bibkey>
-      <pwccode url="https://github.com/benjamin-marie/meta_evaluation_mt" additional="true">benjamin-marie/meta_evaluation_mt</pwccode>
       <video href="2021.acl-long.566.mp4"/>
+      <pwccode url="https://github.com/benjamin-marie/meta_evaluation_mt" additional="true">benjamin-marie/meta_evaluation_mt</pwccode>
     </paper>
     <paper id="567">
       <title>Neural Machine Translation with Monolingual Translation Memory</title>
@@ -9185,12 +9192,12 @@
       <award>Outstanding Paper</award>
       <doi>10.18653/v1/2021.acl-long.569</doi>
       <bibkey>sinha-etal-2021-unnatural</bibkey>
+      <video href="2021.acl-long.569.mp4"/>
       <pwccode url="https://github.com/facebookresearch/UNLU" additional="false">facebookresearch/UNLU</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ocnli">OCNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.acl-long.569.mp4"/>
     </paper>
     <paper id="570">
       <title>Including Signed Languages in Natural Language Processing</title>
@@ -9220,8 +9227,8 @@
       <award>Best Paper</award>
       <doi>10.18653/v1/2021.acl-long.571</doi>
       <bibkey>xu-etal-2021-vocabulary</bibkey>
-      <pwccode url="https://github.com/Jingjing-NLP/VOLT" additional="false">Jingjing-NLP/VOLT</pwccode>
       <video href="2021.acl-long.571.mp4"/>
+      <pwccode url="https://github.com/Jingjing-NLP/VOLT" additional="false">Jingjing-NLP/VOLT</pwccode>
     </paper>
   </volume>
   <volume id="short" ingest-date="2021-07-25">
@@ -9263,8 +9270,8 @@
       <url hash="3d40a31c">2021.acl-short.2</url>
       <doi>10.18653/v1/2021.acl-short.2</doi>
       <bibkey>chang-etal-2021-training</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <video href="2021.acl-short.2.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
     </paper>
     <paper id="3">
       <title>Coreference Resolution without Span Representations</title>
@@ -9276,9 +9283,9 @@
       <url hash="410a03a2">2021.acl-short.3</url>
       <doi>10.18653/v1/2021.acl-short.3</doi>
       <bibkey>kirstain-etal-2021-coreference</bibkey>
+      <video href="2021.acl-short.3.mp4"/>
       <pwccode url="https://github.com/yuvalkirstain/s2e-coref" additional="false">yuvalkirstain/s2e-coref</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gap-coreference-dataset">GAP Coreference Dataset</pwcdataset>
-      <video href="2021.acl-short.3.mp4"/>
     </paper>
     <paper id="4">
       <title>Enhancing Entity Boundary Detection for Better <fixed-case>C</fixed-case>hinese Named Entity Recognition</title>
@@ -9302,8 +9309,8 @@
       <url hash="68956287">2021.acl-short.5</url>
       <doi>10.18653/v1/2021.acl-short.5</doi>
       <bibkey>zhan-etal-2021-difficulty</bibkey>
-      <pwccode url="https://github.com/NLP2CT/Difficulty-Aware-MT-Evaluation" additional="false">NLP2CT/Difficulty-Aware-MT-Evaluation</pwccode>
       <video href="2021.acl-short.5.mp4"/>
+      <pwccode url="https://github.com/NLP2CT/Difficulty-Aware-MT-Evaluation" additional="false">NLP2CT/Difficulty-Aware-MT-Evaluation</pwccode>
     </paper>
     <paper id="6">
       <title>Uncertainty and Surprisal Jointly Deliver the Punchline: Exploiting Incongruity-Based Features for Humor Recognition</title>
@@ -9330,8 +9337,8 @@
       <url hash="970a8c00">2021.acl-short.7</url>
       <doi>10.18653/v1/2021.acl-short.7</doi>
       <bibkey>nangi-etal-2021-counterfactuals</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
       <video href="2021.acl-short.7.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
     </paper>
     <paper id="8">
       <title>Attention Flows are Shapley Value Explanations</title>
@@ -9353,8 +9360,8 @@
       <url hash="e1389264">2021.acl-short.9</url>
       <doi>10.18653/v1/2021.acl-short.9</doi>
       <bibkey>liu-wan-2021-video</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/activitynet-captions">ActivityNet Captions</pwcdataset>
       <video href="2021.acl-short.9.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/activitynet-captions">ActivityNet Captions</pwcdataset>
     </paper>
     <paper id="10">
       <title>Are <fixed-case>VQA</fixed-case> Systems <fixed-case>RAD</fixed-case>? <fixed-case>M</fixed-case>easuring Robustness to Augmented Data with Focused Interventions</title>
@@ -9367,9 +9374,9 @@
       <url hash="50ca2811">2021.acl-short.10</url>
       <doi>10.18653/v1/2021.acl-short.10</doi>
       <bibkey>rosenberg-etal-2021-vqa</bibkey>
+      <video href="2021.acl-short.10.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/visdial">VisDial</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.acl-short.10.mp4"/>
     </paper>
     <paper id="11">
       <title>How Helpful is Inverse Reinforcement Learning for Table-to-Text Generation?</title>
@@ -9383,8 +9390,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="cda68391">2021.acl-short.11.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.11</doi>
       <bibkey>ghosh-etal-2021-helpful</bibkey>
-      <pwccode url="https://github.com/issacqzh/irl_table2text" additional="false">issacqzh/irl_table2text</pwccode>
       <video href="2021.acl-short.11.mp4"/>
+      <pwccode url="https://github.com/issacqzh/irl_table2text" additional="false">issacqzh/irl_table2text</pwccode>
     </paper>
     <paper id="12">
       <title>Automatic Fake News Detection: Are Models Learning to Reason?</title>
@@ -9396,8 +9403,8 @@
       <url hash="e7d84e26">2021.acl-short.12</url>
       <doi>10.18653/v1/2021.acl-short.12</doi>
       <bibkey>hansen-etal-2021-automatic</bibkey>
-      <pwccode url="https://github.com/casperhansen/fake-news-reasoning" additional="false">casperhansen/fake-news-reasoning</pwccode>
       <video href="2021.acl-short.12.mp4"/>
+      <pwccode url="https://github.com/casperhansen/fake-news-reasoning" additional="false">casperhansen/fake-news-reasoning</pwccode>
     </paper>
     <paper id="13">
       <title><fixed-case>S</fixed-case>aying <fixed-case>N</fixed-case>o is <fixed-case>A</fixed-case>n <fixed-case>A</fixed-case>rt: <fixed-case>C</fixed-case>ontextualized <fixed-case>F</fixed-case>allback <fixed-case>R</fixed-case>esponses for <fixed-case>U</fixed-case>nanswerable <fixed-case>D</fixed-case>ialogue <fixed-case>Q</fixed-case>ueries</title>
@@ -9410,8 +9417,8 @@
       <url hash="6ae3d318">2021.acl-short.13</url>
       <doi>10.18653/v1/2021.acl-short.13</doi>
       <bibkey>shrivastava-etal-2021-saying</bibkey>
-      <pwccode url="https://github.com/kaustubhdhole/natural-dont-know" additional="false">kaustubhdhole/natural-dont-know</pwccode>
       <video href="2021.acl-short.13.mp4"/>
+      <pwccode url="https://github.com/kaustubhdhole/natural-dont-know" additional="false">kaustubhdhole/natural-dont-know</pwccode>
     </paper>
     <paper id="14">
       <title>N-Best <fixed-case>ASR</fixed-case> Transformer: Enhancing <fixed-case>SLU</fixed-case> Performance using Multiple <fixed-case>ASR</fixed-case> Hypotheses</title>
@@ -9425,8 +9432,8 @@
       <url hash="705f380c">2021.acl-short.14</url>
       <doi>10.18653/v1/2021.acl-short.14</doi>
       <bibkey>ganesan-etal-2021-n</bibkey>
-      <pwccode url="https://github.com/Vernacular-ai/N-Best-ASR-Transformer" additional="false">Vernacular-ai/N-Best-ASR-Transformer</pwccode>
       <video href="2021.acl-short.14.mp4"/>
+      <pwccode url="https://github.com/Vernacular-ai/N-Best-ASR-Transformer" additional="false">Vernacular-ai/N-Best-ASR-Transformer</pwccode>
     </paper>
     <paper id="15">
       <title>Gender bias amplification during Speed-Quality optimization in Neural Machine Translation</title>
@@ -9454,8 +9461,8 @@
       <url hash="33644c57">2021.acl-short.16</url>
       <doi>10.18653/v1/2021.acl-short.16</doi>
       <bibkey>kumar-etal-2021-machine</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wikimatrix">WikiMatrix</pwcdataset>
       <video href="2021.acl-short.16.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikimatrix">WikiMatrix</pwcdataset>
     </paper>
     <paper id="17">
       <title>Is Sparse Attention more Interpretable?</title>
@@ -9480,10 +9487,10 @@
       <url hash="db40c04d">2021.acl-short.18</url>
       <doi>10.18653/v1/2021.acl-short.18</doi>
       <bibkey>wennberg-henter-2021-case</bibkey>
+      <video href="2021.acl-short.18.mp4"/>
       <pwccode url="https://github.com/ulmewennberg/tisa" additional="false">ulmewennberg/tisa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.acl-short.18.mp4"/>
     </paper>
     <paper id="19">
       <title>Relative Importance in Sentence Processing</title>
@@ -9495,8 +9502,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="389364bb">2021.acl-short.19.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.19</doi>
       <bibkey>hollenstein-beinborn-2021-relative</bibkey>
-      <pwccode url="https://github.com/beinborn/relative_importance" additional="false">beinborn/relative_importance</pwccode>
       <video href="2021.acl-short.19.mp4"/>
+      <pwccode url="https://github.com/beinborn/relative_importance" additional="false">beinborn/relative_importance</pwccode>
     </paper>
     <paper id="20">
       <title>Doing Good or Doing Right? Exploring the Weakness of Commonsense Causal Reasoning Models</title>
@@ -9508,10 +9515,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="f4972a99">2021.acl-short.20.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.20</doi>
       <bibkey>han-wang-2021-good</bibkey>
+      <video href="2021.acl-short.20.mp4"/>
       <pwccode url="https://github.com/badbadcode/weakCOPA" additional="false">badbadcode/weakCOPA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bcopa-ce">BCOPA-CE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
-      <video href="2021.acl-short.20.mp4"/>
     </paper>
     <paper id="21">
       <title><fixed-case>AND</fixed-case> does not mean <fixed-case>OR</fixed-case>: Using Formal Languages to Study Language Models’ Representations</title>
@@ -9535,9 +9542,9 @@
       <url hash="9b00fea4">2021.acl-short.22</url>
       <doi>10.18653/v1/2021.acl-short.22</doi>
       <bibkey>gupta-etal-2021-enforcing</bibkey>
+      <video href="2021.acl-short.22.mp4"/>
       <pwccode url="https://github.com/nitishgupta/allennlp-semparse" additional="false">nitishgupta/allennlp-semparse</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/nlvr">NLVR</pwcdataset>
-      <video href="2021.acl-short.22.mp4"/>
     </paper>
     <paper id="23">
       <title>An Improved Model for Voicing Silent Speech</title>
@@ -9548,8 +9555,8 @@
       <url hash="c44c3125">2021.acl-short.23</url>
       <doi>10.18653/v1/2021.acl-short.23</doi>
       <bibkey>gaddy-klein-2021-improved</bibkey>
-      <pwccode url="https://github.com/dgaddy/silent_speech" additional="false">dgaddy/silent_speech</pwccode>
       <video href="2021.acl-short.23.mp4"/>
+      <pwccode url="https://github.com/dgaddy/silent_speech" additional="false">dgaddy/silent_speech</pwccode>
     </paper>
     <paper id="24">
       <title>What’s in the Box? An Analysis of Undesirable Content in the <fixed-case>C</fixed-case>ommon <fixed-case>C</fixed-case>rawl Corpus</title>
@@ -9599,12 +9606,12 @@
       <attachment type="OptionalSupplementaryMaterial" hash="5c7b2a81">2021.acl-short.27.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.27</doi>
       <bibkey>hessel-schofield-2021-effective</bibkey>
+      <video href="2021.acl-short.27.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-short.27.mp4"/>
     </paper>
     <paper id="28">
       <title><fixed-case>W</fixed-case>iki<fixed-case>S</fixed-case>um: Coherent Summarization Dataset for Efficient Human-Evaluation</title>
@@ -9617,10 +9624,10 @@
       <url hash="aa75cfcb">2021.acl-short.28</url>
       <doi>10.18653/v1/2021.acl-short.28</doi>
       <bibkey>cohen-etal-2021-wikisum</bibkey>
+      <video href="2021.acl-short.28.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bigpatent">BigPatent</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihow">WikiHow</pwcdataset>
-      <video href="2021.acl-short.28.mp4"/>
     </paper>
     <paper id="29">
       <title><fixed-case>UMIC</fixed-case>: An Unreferenced Metric for Image Captioning via Contrastive Learning</title>
@@ -9635,8 +9642,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="fb276b98">2021.acl-short.29.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-short.29</doi>
       <bibkey>lee-etal-2021-umic</bibkey>
-      <pwccode url="https://github.com/hwanheelee1993/UMIC" additional="false">hwanheelee1993/UMIC</pwccode>
       <video href="2021.acl-short.29.mp4"/>
+      <pwccode url="https://github.com/hwanheelee1993/UMIC" additional="false">hwanheelee1993/UMIC</pwccode>
     </paper>
     <paper id="30">
       <title>Anchor-based Bilingual Word Embeddings for Low-Resource Languages</title>
@@ -9689,9 +9696,9 @@
       <url hash="36222a0e">2021.acl-short.33</url>
       <doi>10.18653/v1/2021.acl-short.33</doi>
       <bibkey>yadav-etal-2021-reinforcement</bibkey>
+      <video href="2021.acl-short.33.mp4"/>
       <pwccode url="https://github.com/shwetanlp/CHQ-Summ" additional="false">shwetanlp/CHQ-Summ</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/matinf">MATINF</pwcdataset>
-      <video href="2021.acl-short.33.mp4"/>
     </paper>
     <paper id="34">
       <title>A Semantics-aware Transformer Model of Relation Linking for Knowledge Base Question Answering</title>
@@ -9723,9 +9730,9 @@
       <url hash="794e9b7f">2021.acl-short.35</url>
       <doi>10.18653/v1/2021.acl-short.35</doi>
       <bibkey>yang-etal-2021-neural-retrieval</bibkey>
+      <video href="2021.acl-short.35.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/reqa">ReQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-short.35.mp4"/>
     </paper>
     <paper id="36">
       <title>Enhancing Descriptive Image Captioning with Natural Language Inference</title>
@@ -9737,10 +9744,10 @@
       <url hash="37cc91ca">2021.acl-short.36</url>
       <doi>10.18653/v1/2021.acl-short.36</doi>
       <bibkey>shi-etal-2021-enhancing</bibkey>
+      <video href="2021.acl-short.36.mp4"/>
       <pwccode url="https://github.com/gitsamshi/nli-image-caption" additional="false">gitsamshi/nli-image-caption</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.acl-short.36.mp4"/>
     </paper>
     <paper id="37">
       <title><fixed-case>MOLEMAN</fixed-case>: Mention-Only Linking of Entities with a Mention Annotation Network</title>
@@ -9755,8 +9762,8 @@
       <url hash="d3695656">2021.acl-short.37</url>
       <doi>10.18653/v1/2021.acl-short.37</doi>
       <bibkey>fitzgerald-etal-2021-moleman</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mewsli-9">Mewsli-9</pwcdataset>
       <video href="2021.acl-short.37.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mewsli-9">Mewsli-9</pwcdataset>
     </paper>
     <paper id="38">
       <title>e<fixed-case>MLM</fixed-case>: A New Pre-training Objective for Emotion Related Tasks</title>
@@ -9767,11 +9774,11 @@
       <url hash="a583861a">2021.acl-short.38</url>
       <doi>10.18653/v1/2021.acl-short.38</doi>
       <bibkey>sosea-caragea-2021-emlm</bibkey>
+      <video href="2021.acl-short.38.mp4"/>
       <pwccode url="https://github.com/tsosea2/emlm" additional="false">tsosea2/emlm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/goemotions">GoEmotions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-short.38.mp4"/>
     </paper>
     <paper id="39">
       <title>On Positivity Bias in Negative Reviews</title>
@@ -9782,11 +9789,11 @@
       <url hash="377b96a9">2021.acl-short.39</url>
       <doi>10.18653/v1/2021.acl-short.39</doi>
       <bibkey>aithal-tan-2021-positivity</bibkey>
+      <video href="2021.acl-short.39.mp4"/>
       <pwccode url="https://github.com/madhu-aithal/Positivity-Bias-in-Negative-Reviews" additional="false">madhu-aithal/Positivity-Bias-in-Negative-Reviews</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/peerread">PeerRead</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-short.39.mp4"/>
     </paper>
     <paper id="40">
       <title><fixed-case>PRAL</fixed-case>: A Tailored Pre-Training Model for Task-Oriented Dialog Generation</title>
@@ -9817,8 +9824,8 @@
       <url hash="083d208d">2021.acl-short.41</url>
       <doi>10.18653/v1/2021.acl-short.41</doi>
       <bibkey>lee-etal-2021-rope</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/funsd">FUNSD</pwcdataset>
       <video href="2021.acl-short.41.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/funsd">FUNSD</pwcdataset>
     </paper>
     <paper id="42">
       <title>Zero-shot Event Extraction via Transfer Learning: <fixed-case>C</fixed-case>hallenges and Insights</title>
@@ -9832,8 +9839,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="4d2a2de6">2021.acl-short.42.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.42</doi>
       <bibkey>lyu-etal-2021-zero</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/qamr">QAMR</pwcdataset>
       <video href="2021.acl-short.42.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/qamr">QAMR</pwcdataset>
     </paper>
     <paper id="43">
       <title>Using Adversarial Attacks to Reveal the Statistical Bias in Machine Reading Comprehension Models</title>
@@ -9845,8 +9852,8 @@
       <url hash="3d5bd066">2021.acl-short.43</url>
       <doi>10.18653/v1/2021.acl-short.43</doi>
       <bibkey>lin-etal-2021-using</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
       <video href="2021.acl-short.43.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
     </paper>
     <paper id="44">
       <title>Quantifying and Avoiding Unfair Qualification Labour in Crowdsourcing</title>
@@ -9869,8 +9876,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="e6a19d05">2021.acl-short.45.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.45</doi>
       <bibkey>sun-peng-2021-men</bibkey>
-      <pwccode url="https://github.com/PlusLabNLP/ee-wiki-bias" additional="false">PlusLabNLP/ee-wiki-bias</pwccode>
       <video href="2021.acl-short.45.mp4"/>
+      <pwccode url="https://github.com/PlusLabNLP/ee-wiki-bias" additional="false">PlusLabNLP/ee-wiki-bias</pwccode>
     </paper>
     <paper id="46">
       <title>Modeling Task-Aware <fixed-case>MIMO</fixed-case> Cardinality for Efficient Multilingual Neural Machine Translation</title>
@@ -9883,8 +9890,8 @@
       <url hash="da0710be">2021.acl-short.46</url>
       <doi>10.18653/v1/2021.acl-short.46</doi>
       <bibkey>xu-etal-2021-modeling</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/opus-100">OPUS-100</pwcdataset>
       <video href="2021.acl-short.46.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/opus-100">OPUS-100</pwcdataset>
     </paper>
     <paper id="47">
       <title>Adaptive Nearest Neighbor Machine Translation</title>
@@ -9900,8 +9907,8 @@
       <url hash="24f9f4e4">2021.acl-short.47</url>
       <doi>10.18653/v1/2021.acl-short.47</doi>
       <bibkey>zheng-etal-2021-adaptive</bibkey>
-      <pwccode url="https://github.com/zhengxxn/adaptive-knn-mt" additional="false">zhengxxn/adaptive-knn-mt</pwccode>
       <video href="2021.acl-short.47.mp4"/>
+      <pwccode url="https://github.com/zhengxxn/adaptive-knn-mt" additional="false">zhengxxn/adaptive-knn-mt</pwccode>
     </paper>
     <paper id="48">
       <title>On Orthogonality Constraints for Transformers</title>
@@ -9919,9 +9926,9 @@
       <url hash="033e95fe">2021.acl-short.48</url>
       <doi>10.18653/v1/2021.acl-short.48</doi>
       <bibkey>zhang-etal-2021-orthogonality</bibkey>
+      <video href="2021.acl-short.48.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.acl-short.48.mp4"/>
     </paper>
     <paper id="49">
       <title>Measuring and Improving <fixed-case>BERT</fixed-case>’s Mathematical Abilities by Predicting the Order of Reasoning.</title>
@@ -9934,8 +9941,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="b506ee1f">2021.acl-short.49.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.49</doi>
       <bibkey>piekos-etal-2021-measuring</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/aqua-rat">AQUA-RAT</pwcdataset>
       <video href="2021.acl-short.49.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/aqua-rat">AQUA-RAT</pwcdataset>
     </paper>
     <paper id="50">
       <title>Happy Dance, Slow Clap: <fixed-case>Using</fixed-case> Reaction <fixed-case>GIFs</fixed-case> to Predict Induced Affect on <fixed-case>Twitter</fixed-case></title>
@@ -9947,8 +9954,8 @@
       <url hash="8fb92c3a">2021.acl-short.50</url>
       <doi>10.18653/v1/2021.acl-short.50</doi>
       <bibkey>shmueli-etal-2021-happy</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/reactiongif">ReactionGIF</pwcdataset>
       <video href="2021.acl-short.50.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/reactiongif">ReactionGIF</pwcdataset>
     </paper>
     <paper id="51">
       <title>Exploring Listwise Evidence Reasoning with T5 for Fact Verification</title>
@@ -9961,8 +9968,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="cc14c701">2021.acl-short.51.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.51</doi>
       <bibkey>jiang-etal-2021-exploring-listwise</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <video href="2021.acl-short.51.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
     </paper>
     <paper id="52">
       <title><fixed-case>D</fixed-case>ef<fixed-case>S</fixed-case>ent: Sentence Embeddings using Definition Sentences</title>
@@ -9974,9 +9981,9 @@
       <url hash="c8c75ea7">2021.acl-short.52</url>
       <doi>10.18653/v1/2021.acl-short.52</doi>
       <bibkey>tsukagoshi-etal-2021-defsent</bibkey>
+      <video href="2021.acl-short.52.mp4"/>
       <pwccode url="https://github.com/hppRC/defsent" additional="false">hppRC/defsent</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.acl-short.52.mp4"/>
     </paper>
     <paper id="53">
       <title>Discrete Cosine Transform as Universal Sentence Encoder</title>
@@ -9999,8 +10006,8 @@
       <url hash="46c09b4d">2021.acl-short.54</url>
       <doi>10.18653/v1/2021.acl-short.54</doi>
       <bibkey>mirza-etal-2021-alignarr</bibkey>
-      <pwccode url="https://github.com/paramitamirza/alignarr" additional="false">paramitamirza/alignarr</pwccode>
       <video href="2021.acl-short.54.mp4"/>
+      <pwccode url="https://github.com/paramitamirza/alignarr" additional="false">paramitamirza/alignarr</pwccode>
     </paper>
     <paper id="55">
       <title>An Exploratory Analysis of Multilingual Word-Level Quality Estimation with Cross-Lingual Transformers</title>
@@ -10013,8 +10020,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="3eb9de5b">2021.acl-short.55.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-short.55</doi>
       <bibkey>ranasinghe-etal-2021-exploratory</bibkey>
-      <pwccode url="https://github.com/tharindudr/transQuest" additional="false">tharindudr/transQuest</pwccode>
       <video href="2021.acl-short.55.mp4"/>
+      <pwccode url="https://github.com/tharindudr/transQuest" additional="false">tharindudr/transQuest</pwccode>
     </paper>
     <paper id="56">
       <title>Exploration and Exploitation: Two Ways to Improve <fixed-case>C</fixed-case>hinese Spelling Correction Models</title>
@@ -10028,8 +10035,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="11c907b5">2021.acl-short.56.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.56</doi>
       <bibkey>li-etal-2021-exploration</bibkey>
-      <pwccode url="https://github.com/FDChongli/TwoWaysToImproveCSC" additional="false">FDChongli/TwoWaysToImproveCSC</pwccode>
       <video href="2021.acl-short.56.mp4"/>
+      <pwccode url="https://github.com/FDChongli/TwoWaysToImproveCSC" additional="false">FDChongli/TwoWaysToImproveCSC</pwccode>
     </paper>
     <paper id="57">
       <title>Training Adaptive Computation for Open-Domain Question Answering with Computational Constraints</title>
@@ -10042,10 +10049,10 @@
       <url hash="98f2fc61">2021.acl-short.57</url>
       <doi>10.18653/v1/2021.acl-short.57</doi>
       <bibkey>wu-etal-2021-training</bibkey>
+      <video href="2021.acl-short.57.mp4"/>
       <pwccode url="https://github.com/uclnlp/APE" additional="false">uclnlp/APE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.acl-short.57.mp4"/>
     </paper>
     <paper id="58">
       <title>An Empirical Study on Adversarial Attack on <fixed-case>NMT</fixed-case>: Languages and Positions Matter</title>
@@ -10068,12 +10075,12 @@
       <url hash="f4c2995e">2021.acl-short.59</url>
       <doi>10.18653/v1/2021.acl-short.59</doi>
       <bibkey>zhu-etal-2021-ontogum</bibkey>
+      <video href="2021.acl-short.59.mp4"/>
       <pwccode url="https://github.com/yilunzhu/ontogum" additional="false">yilunzhu/ontogum</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ontogum">OntoGUM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gap-coreference-dataset">GAP Coreference Dataset</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gum">GUM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikicoref">WikiCoref</pwcdataset>
-      <video href="2021.acl-short.59.mp4"/>
     </paper>
     <paper id="60">
       <title>In Factuality: Efficient Integration of Relevant Facts for Visual Question Answering</title>
@@ -10086,8 +10093,8 @@
       <url hash="357a7ec4">2021.acl-short.60</url>
       <doi>10.18653/v1/2021.acl-short.60</doi>
       <bibkey>vickers-etal-2021-factuality</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
       <video href="2021.acl-short.60.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
     </paper>
     <paper id="61">
       <title>Zero-shot Fact Verification by Claim Generation</title>
@@ -10101,10 +10108,10 @@
       <url hash="e6ce66ce">2021.acl-short.61</url>
       <doi>10.18653/v1/2021.acl-short.61</doi>
       <bibkey>pan-etal-2021-zero</bibkey>
+      <video href="2021.acl-short.61.mp4"/>
       <pwccode url="https://github.com/teacherpeterpan/Zero-shot-Fact-Verification" additional="false">teacherpeterpan/Zero-shot-Fact-Verification</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qa2d">QA2D</pwcdataset>
-      <video href="2021.acl-short.61.mp4"/>
     </paper>
     <paper id="62">
       <title>Thank you <fixed-case>BART</fixed-case>! Rewarding Pre-Trained Models Improves Formality Style Transfer</title>
@@ -10116,9 +10123,9 @@
       <url hash="6866fcea">2021.acl-short.62</url>
       <doi>10.18653/v1/2021.acl-short.62</doi>
       <bibkey>lai-etal-2021-thank</bibkey>
+      <video href="2021.acl-short.62.mp4"/>
       <pwccode url="https://github.com/laihuiyuan/Pre-trained-formality-transfer" additional="false">laihuiyuan/Pre-trained-formality-transfer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
-      <video href="2021.acl-short.62.mp4"/>
     </paper>
     <paper id="63">
       <title>Deep Context- and Relation-Aware Learning for Aspect-based Sentiment Analysis</title>
@@ -10148,9 +10155,9 @@
       <url hash="55f811db">2021.acl-short.64</url>
       <doi>10.18653/v1/2021.acl-short.64</doi>
       <bibkey>zhang-etal-2021-towards-generative</bibkey>
+      <video href="2021.acl-short.64.mp4"/>
       <pwccode url="https://github.com/IsakZhang/Generative-ABSA" additional="false">IsakZhang/Generative-ABSA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/aste-data-v2">ASTE-Data-V2</pwcdataset>
-      <video href="2021.acl-short.64.mp4"/>
     </paper>
     <paper id="65">
       <title>Bilingual Mutual Information Based Adaptive Training for Neural Machine Translation</title>
@@ -10166,8 +10173,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="101d8c88">2021.acl-short.65.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.65</doi>
       <bibkey>xu-etal-2021-bilingual</bibkey>
-      <pwccode url="https://github.com/xydaytoy/BMI-NMT" additional="false">xydaytoy/BMI-NMT</pwccode>
       <video href="2021.acl-short.65.mp4"/>
+      <pwccode url="https://github.com/xydaytoy/BMI-NMT" additional="false">xydaytoy/BMI-NMT</pwccode>
     </paper>
     <paper id="66">
       <title>Continual Learning for Task-oriented Dialogue System with Iterative Network Pruning, Expanding and Masking</title>
@@ -10182,8 +10189,8 @@
       <url hash="ea0b90c3">2021.acl-short.66</url>
       <doi>10.18653/v1/2021.acl-short.66</doi>
       <bibkey>geng-etal-2021-continual</bibkey>
-      <pwccode url="https://github.com/siat-nlp/TPEM" additional="false">siat-nlp/TPEM</pwccode>
       <video href="2021.acl-short.66.mp4"/>
+      <pwccode url="https://github.com/siat-nlp/TPEM" additional="false">siat-nlp/TPEM</pwccode>
     </paper>
     <paper id="67">
       <title><fixed-case>TIMERS</fixed-case>: Document-level Temporal Relation Extraction</title>
@@ -10211,8 +10218,8 @@
       <url hash="5c3bc916">2021.acl-short.68</url>
       <doi>10.18653/v1/2021.acl-short.68</doi>
       <bibkey>qin-etal-2021-improving</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/arabic-text-diacritization">Arabic Text Diacritization</pwcdataset>
       <video href="2021.acl-short.68.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/arabic-text-diacritization">Arabic Text Diacritization</pwcdataset>
     </paper>
     <paper id="69">
       <title>When is Char Better Than Subword: A Systematic Study of Segmentation Algorithms for Neural Machine Translation</title>
@@ -10241,8 +10248,8 @@
       <url hash="0034d4a0">2021.acl-short.70</url>
       <doi>10.18653/v1/2021.acl-short.70</doi>
       <bibkey>zhang-etal-2021-text</bibkey>
-      <pwccode url="https://github.com/manlp-suda/mcws" additional="false">manlp-suda/mcws</pwccode>
       <video href="2021.acl-short.70.mp4"/>
+      <pwccode url="https://github.com/manlp-suda/mcws" additional="false">manlp-suda/mcws</pwccode>
     </paper>
     <paper id="71">
       <title>A Mixture-of-Experts Model for Antonym-Synonym Discrimination</title>
@@ -10254,8 +10261,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="99d9c3f8">2021.acl-short.71.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.71</doi>
       <bibkey>xie-zeng-2021-mixture</bibkey>
-      <pwccode url="https://github.com/zengnan1997/moe-asd" additional="false">zengnan1997/moe-asd</pwccode>
       <video href="2021.acl-short.71.mp4"/>
+      <pwccode url="https://github.com/zengnan1997/moe-asd" additional="false">zengnan1997/moe-asd</pwccode>
     </paper>
     <paper id="72">
       <title>Learning Domain-Specialised Representations for Cross-Lingual Biomedical Entity Linking</title>
@@ -10268,9 +10275,9 @@
       <url hash="7131618f">2021.acl-short.72</url>
       <doi>10.18653/v1/2021.acl-short.72</doi>
       <bibkey>liu-etal-2021-learning-domain</bibkey>
+      <video href="2021.acl-short.72.mp4"/>
       <pwccode url="https://github.com/cambridgeltl/sapbert" additional="false">cambridgeltl/sapbert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/xl-bel">XL-BEL</pwcdataset>
-      <video href="2021.acl-short.72.mp4"/>
     </paper>
     <paper id="73">
       <title>A Cluster-based Approach for Improving Isotropy in Contextual Embedding Space</title>
@@ -10281,6 +10288,7 @@
       <url hash="c3c76a7b">2021.acl-short.73</url>
       <doi>10.18653/v1/2021.acl-short.73</doi>
       <bibkey>rajaee-pilehvar-2021-cluster</bibkey>
+      <video href="2021.acl-short.73.mp4"/>
       <pwccode url="https://github.com/Sara-Rajaee/clusterbased_isotropy_enhancement" additional="false">Sara-Rajaee/clusterbased_isotropy_enhancement</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
@@ -10289,7 +10297,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sts-2014">STS 2014</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
-      <video href="2021.acl-short.73.mp4"/>
     </paper>
     <paper id="74">
       <title>Unsupervised Enrichment of Persona-grounded Dialog with Background Stories</title>
@@ -10302,9 +10309,9 @@
       <url hash="d5a67b64">2021.acl-short.74</url>
       <doi>10.18653/v1/2021.acl-short.74</doi>
       <bibkey>majumder-etal-2021-unsupervised</bibkey>
+      <video href="2021.acl-short.74.mp4"/>
       <pwccode url="https://github.com/majumderb/pabst" additional="false">majumderb/pabst</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.acl-short.74.mp4"/>
     </paper>
     <paper id="75">
       <title>Beyond Laurel/Yanny: An Autoencoder-Enabled Search for Polyperceivable Audio</title>
@@ -10341,9 +10348,9 @@
       <url hash="2e4f685a">2021.acl-short.77</url>
       <doi>10.18653/v1/2021.acl-short.77</doi>
       <bibkey>reimers-gurevych-2021-curse</bibkey>
+      <video href="2021.acl-short.77.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.acl-short.77.mp4"/>
     </paper>
     <paper id="78">
       <title>Cross-lingual Text Classification with Heterogeneous Graph Neural Network</title>
@@ -10357,9 +10364,9 @@
       <url hash="2c33e7ff">2021.acl-short.78</url>
       <doi>10.18653/v1/2021.acl-short.78</doi>
       <bibkey>wang-etal-2021-cross-lingual</bibkey>
+      <video href="2021.acl-short.78.mp4"/>
       <pwccode url="https://github.com/TencentGameMate/gnn_cross_lingual" additional="false">TencentGameMate/gnn_cross_lingual</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/xglue">XGLUE</pwcdataset>
-      <video href="2021.acl-short.78.mp4"/>
     </paper>
     <paper id="79">
       <title>Towards more equitable question answering systems: How much more data do you need?</title>
@@ -10372,11 +10379,11 @@
       <url hash="b75b41e3">2021.acl-short.79</url>
       <doi>10.18653/v1/2021.acl-short.79</doi>
       <bibkey>debnath-etal-2021-towards</bibkey>
+      <video href="2021.acl-short.79.mp4"/>
       <pwccode url="https://github.com/NavidRajabi/EMQA" additional="false">NavidRajabi/EMQA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
-      <video href="2021.acl-short.79.mp4"/>
     </paper>
     <paper id="80">
       <title>Embedding Time Differences in Context-sensitive Neural Networks for Learning Time to Event</title>
@@ -10401,8 +10408,8 @@
       <url hash="5e8acb47">2021.acl-short.81</url>
       <doi>10.18653/v1/2021.acl-short.81</doi>
       <bibkey>kim-etal-2021-improving</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
       <video href="2021.acl-short.81.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
     </paper>
     <paper id="82">
       <title>Learning to Generate Task-Specific Adapters from Task Description</title>
@@ -10414,12 +10421,12 @@
       <attachment type="OptionalSupplementaryMaterial" hash="c0e69651">2021.acl-short.82.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.82</doi>
       <bibkey>ye-ren-2021-learning</bibkey>
+      <video href="2021.acl-short.82.mp4"/>
       <pwccode url="https://github.com/INK-USC/hypter" additional="false">INK-USC/hypter</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsqa">NewsQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/zest">ZEST</pwcdataset>
-      <video href="2021.acl-short.82.mp4"/>
     </paper>
     <paper id="83">
       <title><fixed-case>QA</fixed-case>-Driven Zero-shot Slot Filling with Weak Supervision Pretraining</title>
@@ -10434,8 +10441,8 @@
       <url hash="2fd329b5">2021.acl-short.83</url>
       <doi>10.18653/v1/2021.acl-short.83</doi>
       <bibkey>du-etal-2021-qa</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
       <video href="2021.acl-short.83.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
     </paper>
     <paper id="84">
       <title>Domain-Adaptive Pretraining Methods for Dialogue Understanding</title>
@@ -10450,8 +10457,8 @@
       <url hash="fdc46848">2021.acl-short.84</url>
       <doi>10.18653/v1/2021.acl-short.84</doi>
       <bibkey>wu-etal-2021-domain</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/crosswoz">CrossWOZ</pwcdataset>
       <video href="2021.acl-short.84.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/crosswoz">CrossWOZ</pwcdataset>
     </paper>
     <paper id="85">
       <title>Targeting the Benchmark: On Methodology in Current Natural Language Processing Research</title>
@@ -10461,8 +10468,8 @@
       <url hash="fdce1f95">2021.acl-short.85</url>
       <doi>10.18653/v1/2021.acl-short.85</doi>
       <bibkey>schlangen-2021-targeting</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
       <video href="2021.acl-short.85.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
     </paper>
     <paper id="86">
       <title><fixed-case>X</fixed-case>-Fact: A New Benchmark Dataset for Multilingual Fact Checking</title>
@@ -10473,9 +10480,9 @@
       <url hash="95e1de6c">2021.acl-short.86</url>
       <doi>10.18653/v1/2021.acl-short.86</doi>
       <bibkey>gupta-srikumar-2021-x</bibkey>
+      <video href="2021.acl-short.86.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/x-fact">X-Fact</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multifc">MultiFC</pwcdataset>
-      <video href="2021.acl-short.86.mp4"/>
     </paper>
     <paper id="87">
       <title>nm<fixed-case>T</fixed-case>5 - Is parallel data still relevant for pre-training massively multilingual language models?</title>
@@ -10490,9 +10497,9 @@
       <url hash="962e31b0">2021.acl-short.87</url>
       <doi>10.18653/v1/2021.acl-short.87</doi>
       <bibkey>kale-etal-2021-nmt5</bibkey>
+      <video href="2021.acl-short.87.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/opus-100">OPUS-100</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mc4">mC4</pwcdataset>
-      <video href="2021.acl-short.87.mp4"/>
     </paper>
     <paper id="88">
       <title>Question Generation for Adaptive Education</title>
@@ -10503,8 +10510,8 @@
       <url hash="3bae0421">2021.acl-short.88</url>
       <doi>10.18653/v1/2021.acl-short.88</doi>
       <bibkey>srivastava-goodman-2021-question</bibkey>
-      <pwccode url="https://github.com/meghabyte/acl2021-education" additional="false">meghabyte/acl2021-education</pwccode>
       <video href="2021.acl-short.88.mp4"/>
+      <pwccode url="https://github.com/meghabyte/acl2021-education" additional="false">meghabyte/acl2021-education</pwccode>
     </paper>
     <paper id="89">
       <title>A Simple Recipe for Multilingual Grammatical Error Correction</title>
@@ -10518,13 +10525,13 @@
       <url hash="114b4aa8">2021.acl-short.89</url>
       <doi>10.18653/v1/2021.acl-short.89</doi>
       <bibkey>rothe-etal-2021-simple</bibkey>
+      <video href="2021.acl-short.89.mp4"/>
       <pwccode url="https://github.com/google-research-datasets/clang8" additional="false">google-research-datasets/clang8</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/akces-gec">AKCES-GEC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2014-shared-task-grammatical-error">CoNLL-2014 Shared Task: Grammatical Error Correction</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fce">FCE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mc4">mC4</pwcdataset>
-      <video href="2021.acl-short.89.mp4"/>
     </paper>
     <paper id="90">
       <title>Towards Visual Question Answering on Pathology Images</title>
@@ -10540,9 +10547,9 @@
       <url hash="46eb75ed">2021.acl-short.90</url>
       <doi>10.18653/v1/2021.acl-short.90</doi>
       <bibkey>he-etal-2021-towards</bibkey>
+      <video href="2021.acl-short.90.mp4"/>
       <pwccode url="https://github.com/UCSD-AI4H/PathVQA" additional="false">UCSD-AI4H/PathVQA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.acl-short.90.mp4"/>
     </paper>
     <paper id="91">
       <title>Efficient Text-based Reinforcement Learning by Jointly Leveraging State and Commonsense Graph Representations</title>
@@ -10557,8 +10564,8 @@
       <url hash="8d23a14f">2021.acl-short.91</url>
       <doi>10.18653/v1/2021.acl-short.91</doi>
       <bibkey>murugesan-etal-2021-efficient</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <video href="2021.acl-short.91.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
     </paper>
     <paper id="92">
       <title>m<fixed-case>TVR</fixed-case>: Multilingual Moment Retrieval in Videos</title>
@@ -10570,10 +10577,10 @@
       <url hash="cf9d25cc">2021.acl-short.92</url>
       <doi>10.18653/v1/2021.acl-short.92</doi>
       <bibkey>lei-etal-2021-mtvr</bibkey>
+      <video href="2021.acl-short.92.mp4"/>
       <pwccode url="https://github.com/jayleicn/mTVRetrieval" additional="false">jayleicn/mTVRetrieval</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mtvr">mTVR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tvr">TVR</pwcdataset>
-      <video href="2021.acl-short.92.mp4"/>
     </paper>
     <paper id="93">
       <title>Explicitly Capturing Relations between Entity Mentions via Graph Neural Networks for Domain-specific Named Entity Recognition</title>
@@ -10586,8 +10593,8 @@
       <url hash="7574f091">2021.acl-short.93</url>
       <doi>10.18653/v1/2021.acl-short.93</doi>
       <bibkey>chen-etal-2021-explicitly</bibkey>
-      <pwccode url="https://github.com/brickee/enrel-g" additional="false">brickee/enrel-g</pwccode>
       <video href="2021.acl-short.93.mp4"/>
+      <pwccode url="https://github.com/brickee/enrel-g" additional="false">brickee/enrel-g</pwccode>
     </paper>
     <paper id="94">
       <title>Improving Lexically Constrained Neural Machine Translation with Source-Conditioned Masked Span Prediction</title>
@@ -10599,8 +10606,8 @@
       <url hash="1f49a697">2021.acl-short.94</url>
       <doi>10.18653/v1/2021.acl-short.94</doi>
       <bibkey>lee-etal-2021-improving</bibkey>
-      <pwccode url="https://github.com/wns823/NMT_SSP" additional="false">wns823/NMT_SSP</pwccode>
       <video href="2021.acl-short.94.mp4"/>
+      <pwccode url="https://github.com/wns823/NMT_SSP" additional="false">wns823/NMT_SSP</pwccode>
     </paper>
     <paper id="95">
       <title>Quotation Recommendation and Interpretation Based on Transformation from Queries to Quotations</title>
@@ -10612,8 +10619,8 @@
       <url hash="df392dd2">2021.acl-short.95</url>
       <doi>10.18653/v1/2021.acl-short.95</doi>
       <bibkey>wang-etal-2021-quotation</bibkey>
-      <pwccode url="https://github.com/Lingzhi-WANG/Quotation-Recommendation" additional="false">Lingzhi-WANG/Quotation-Recommendation</pwccode>
       <video href="2021.acl-short.95.mp4"/>
+      <pwccode url="https://github.com/Lingzhi-WANG/Quotation-Recommendation" additional="false">Lingzhi-WANG/Quotation-Recommendation</pwccode>
     </paper>
     <paper id="96">
       <title>Pre-training is a Hot Topic: Contextualized Document Embeddings Improve Topic Coherence</title>
@@ -10625,9 +10632,9 @@
       <url hash="b8ebcf67">2021.acl-short.96</url>
       <doi>10.18653/v1/2021.acl-short.96</doi>
       <bibkey>bianchi-etal-2021-pre</bibkey>
+      <video href="2021.acl-short.96.mp4"/>
       <pwccode url="https://github.com/MilaNLProc/contextualized-topic-models" additional="true">MilaNLProc/contextualized-topic-models</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/20-newsgroups">20 Newsgroups</pwcdataset>
-      <video href="2021.acl-short.96.mp4"/>
     </paper>
     <paper id="97">
       <title>Input Representations for Parsing Discourse Representation Structures: Comparing <fixed-case>E</fixed-case>nglish with <fixed-case>C</fixed-case>hinese</title>
@@ -10640,8 +10647,8 @@
       <url hash="7f076cbc">2021.acl-short.97</url>
       <doi>10.18653/v1/2021.acl-short.97</doi>
       <bibkey>wang-etal-2021-input</bibkey>
-      <pwccode url="https://github.com/wangchunliu/chinese-drs-parsing" additional="false">wangchunliu/chinese-drs-parsing</pwccode>
       <video href="2021.acl-short.97.mp4"/>
+      <pwccode url="https://github.com/wangchunliu/chinese-drs-parsing" additional="false">wangchunliu/chinese-drs-parsing</pwccode>
     </paper>
     <paper id="98">
       <title>Code Generation from Natural Language with Less Prior Knowledge and More Monolingual Data</title>
@@ -10653,10 +10660,10 @@
       <url hash="dcbb6303">2021.acl-short.98</url>
       <doi>10.18653/v1/2021.acl-short.98</doi>
       <bibkey>norouzi-etal-2021-code</bibkey>
+      <video href="2021.acl-short.98.mp4"/>
       <pwccode url="https://github.com/borealisai/code-gen-tae" additional="false">borealisai/code-gen-tae</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conala">CoNaLa</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
-      <video href="2021.acl-short.98.mp4"/>
     </paper>
     <paper id="99">
       <title>Issues with Entailment-based Zero-shot Text Classification</title>
@@ -10670,6 +10677,7 @@
       <attachment type="OptionalSupplementaryMaterial" hash="370cb2b6">2021.acl-short.99.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.99</doi>
       <bibkey>ma-etal-2021-issues</bibkey>
+      <video href="2021.acl-short.99.mp4"/>
       <pwccode url="https://github.com/mtt1998/issues-nli" additional="false">mtt1998/issues-nli</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
@@ -10677,7 +10685,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.acl-short.99.mp4"/>
     </paper>
     <paper id="100">
       <title>Neural-Symbolic Commonsense Reasoner with Relation Predictors</title>
@@ -10691,8 +10698,8 @@
       <url hash="13361eed">2021.acl-short.100</url>
       <doi>10.18653/v1/2021.acl-short.100</doi>
       <bibkey>moghimifar-etal-2021-neural</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <video href="2021.acl-short.100.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
     </paper>
     <paper id="101">
       <title>What Motivates You? Benchmarking Automatic Detection of Basic Needs from Short Posts</title>
@@ -10717,8 +10724,8 @@
       <url hash="2bca5ad7">2021.acl-short.102</url>
       <doi>10.18653/v1/2021.acl-short.102</doi>
       <bibkey>yamada-etal-2021-semantic</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
       <video href="2021.acl-short.102.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
     </paper>
     <paper id="103">
       <title>Lightweight Adapter Tuning for Multilingual Speech Translation</title>
@@ -10733,9 +10740,9 @@
       <url hash="28aabd7b">2021.acl-short.103</url>
       <doi>10.18653/v1/2021.acl-short.103</doi>
       <bibkey>le-etal-2021-lightweight</bibkey>
+      <video href="2021.acl-short.103.mp4"/>
       <pwccode url="https://github.com/formiel/fairseq/blob/master/examples/speech_to_text/docs/adapters.md" additional="false">formiel/fairseq</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
-      <video href="2021.acl-short.103.mp4"/>
     </paper>
     <paper id="104">
       <title>Parameter Selection: Why We Should Pay More Attention to It</title>
@@ -10748,8 +10755,8 @@
       <url hash="3cba7ecc">2021.acl-short.104</url>
       <doi>10.18653/v1/2021.acl-short.104</doi>
       <bibkey>liu-etal-2021-parameter</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mimic-iii">MIMIC-III</pwcdataset>
       <video href="2021.acl-short.104.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mimic-iii">MIMIC-III</pwcdataset>
     </paper>
     <paper id="105">
       <title>Distinct Label Representations for Few-Shot Text Classification</title>
@@ -10762,9 +10769,9 @@
       <url hash="61872e82">2021.acl-short.105</url>
       <doi>10.18653/v1/2021.acl-short.105</doi>
       <bibkey>ohashi-etal-2021-distinct</bibkey>
+      <video href="2021.acl-short.105.mp4"/>
       <pwccode url="https://github.com/21335732529sky/difference_extractor" additional="false">21335732529sky/difference_extractor</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
-      <video href="2021.acl-short.105.mp4"/>
     </paper>
     <paper id="106">
       <title>Learning to Solve <fixed-case>NLP</fixed-case> Tasks in an Incremental Number of Languages</title>
@@ -10777,8 +10784,8 @@
       <url hash="f88c733a">2021.acl-short.106</url>
       <doi>10.18653/v1/2021.acl-short.106</doi>
       <bibkey>castellucci-etal-2021-learning</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <video href="2021.acl-short.106.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
     </paper>
     <paper id="107">
       <title>Hi-Transformer: Hierarchical Interactive Transformer for Efficient and Effective Long Document Modeling</title>
@@ -10791,8 +10798,8 @@
       <url hash="1b6742ca">2021.acl-short.107</url>
       <doi>10.18653/v1/2021.acl-short.107</doi>
       <bibkey>wu-etal-2021-hi</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
       <video href="2021.acl-short.107.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
     </paper>
     <paper id="108">
       <title>Robust Transfer Learning with Pretrained Language Models through Adapters</title>
@@ -10804,8 +10811,8 @@
       <url hash="8fc933c8">2021.acl-short.108</url>
       <doi>10.18653/v1/2021.acl-short.108</doi>
       <bibkey>han-etal-2021-robust</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <video href="2021.acl-short.108.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
     </paper>
     <paper id="109">
       <title>Embracing Ambiguity: <fixed-case>S</fixed-case>hifting the Training Target of <fixed-case>NLI</fixed-case> Models</title>
@@ -10818,11 +10825,11 @@
       <url hash="43a58ea9">2021.acl-short.109</url>
       <doi>10.18653/v1/2021.acl-short.109</doi>
       <bibkey>meissner-etal-2021-embracing</bibkey>
+      <video href="2021.acl-short.109.mp4"/>
       <pwccode url="https://github.com/mariomeissner/AmbiNLI" additional="false">mariomeissner/AmbiNLI</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/chaosnli">ChaosNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.acl-short.109.mp4"/>
     </paper>
     <paper id="110">
       <title>Modeling Discriminative Representations for Out-of-Domain Detection with Supervised Contrastive Learning</title>
@@ -10839,8 +10846,8 @@
       <url hash="14416836">2021.acl-short.110</url>
       <doi>10.18653/v1/2021.acl-short.110</doi>
       <bibkey>zeng-etal-2021-modeling</bibkey>
-      <pwccode url="https://github.com/parZival27/supervised-contrastive-learning-for-out-of-domain-detection" additional="false">parZival27/supervised-contrastive-learning-for-out-of-domain-detection</pwccode>
       <video href="2021.acl-short.110.mp4"/>
+      <pwccode url="https://github.com/parZival27/supervised-contrastive-learning-for-out-of-domain-detection" additional="false">parZival27/supervised-contrastive-learning-for-out-of-domain-detection</pwccode>
     </paper>
     <paper id="111">
       <title>Preview, Attend and Review: Schema-Aware Curriculum Learning for Multi-Domain Dialogue State Tracking</title>
@@ -10882,8 +10889,8 @@
       <url hash="2a7f9b48">2021.acl-short.112</url>
       <doi>10.18653/v1/2021.acl-short.112</doi>
       <bibkey>zhou-etal-2021-generation</bibkey>
-      <pwccode url="https://github.com/UCSD-AI4H/COVID-Dialogue" additional="false">UCSD-AI4H/COVID-Dialogue</pwccode>
       <video href="2021.acl-short.112.mp4"/>
+      <pwccode url="https://github.com/UCSD-AI4H/COVID-Dialogue" additional="false">UCSD-AI4H/COVID-Dialogue</pwccode>
     </paper>
     <paper id="113">
       <title>Constructing Multi-Modal Dialogue Dataset by Replacing Text with Semantically Relevant Images</title>
@@ -10897,8 +10904,8 @@
       <url hash="35b50a92">2021.acl-short.113</url>
       <doi>10.18653/v1/2021.acl-short.113</doi>
       <bibkey>lee-etal-2021-constructing</bibkey>
-      <pwccode url="https://github.com/shh1574/multi-modal-dialogue-dataset" additional="false">shh1574/multi-modal-dialogue-dataset</pwccode>
       <video href="2021.acl-short.113.mp4"/>
+      <pwccode url="https://github.com/shh1574/multi-modal-dialogue-dataset" additional="false">shh1574/multi-modal-dialogue-dataset</pwccode>
     </paper>
     <paper id="114">
       <title>Exposing the limits of Zero-shot Cross-lingual Hate Speech Detection</title>
@@ -10943,13 +10950,13 @@
       <url hash="54cc11f2">2021.acl-short.117</url>
       <doi>10.18653/v1/2021.acl-short.117</doi>
       <bibkey>shen-etal-2021-unsupervised</bibkey>
+      <video href="2021.acl-short.117.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/definite-pronoun-resolution-dataset">Definite Pronoun Resolution Dataset</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gap-coreference-dataset">GAP Coreference Dataset</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/pg-19">PG-19</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winogrande">WinoGrande</pwcdataset>
-      <video href="2021.acl-short.117.mp4"/>
     </paper>
     <paper id="118">
       <title>Addressing Semantic Drift in Generative Question Answering with Auxiliary Extraction</title>
@@ -10963,8 +10970,8 @@
       <url hash="ead950a7">2021.acl-short.118</url>
       <doi>10.18653/v1/2021.acl-short.118</doi>
       <bibkey>li-etal-2021-addressing-semantic</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <video href="2021.acl-short.118.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
     </paper>
     <paper id="119">
       <title>Demoting the Lead Bias in News Summarization via Alternating Adversarial Learning</title>
@@ -10991,9 +10998,9 @@
       <url hash="d30dcdcf">2021.acl-short.120</url>
       <doi>10.18653/v1/2021.acl-short.120</doi>
       <bibkey>tang-etal-2021-dureader</bibkey>
+      <video href="2021.acl-short.120.mp4"/>
       <pwccode url="https://github.com/baidu/DuReader" additional="false">baidu/DuReader</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dureader">DuReader</pwcdataset>
-      <video href="2021.acl-short.120.mp4"/>
     </paper>
     <paper id="121">
       <title>Sequence to General Tree: Knowledge-Guided Geometry Word Problem Solving</title>
@@ -11006,8 +11013,8 @@
       <url hash="92c10639">2021.acl-short.121</url>
       <doi>10.18653/v1/2021.acl-short.121</doi>
       <bibkey>tsai-etal-2021-sequence</bibkey>
-      <pwccode url="https://github.com/doublebite/Sequence-to-General-tree" additional="false">doublebite/Sequence-to-General-tree</pwccode>
       <video href="2021.acl-short.121.mp4"/>
+      <pwccode url="https://github.com/doublebite/Sequence-to-General-tree" additional="false">doublebite/Sequence-to-General-tree</pwccode>
     </paper>
     <paper id="122">
       <title>Multi-Scale Progressive Attention Network for Video Question Answering</title>
@@ -11034,11 +11041,11 @@
       <url hash="76c8f71e">2021.acl-short.123</url>
       <doi>10.18653/v1/2021.acl-short.123</doi>
       <bibkey>yamada-etal-2021-efficient</bibkey>
+      <video href="2021.acl-short.123.mp4"/>
       <pwccode url="https://github.com/studio-ousia/bpr" additional="false">studio-ousia/bpr</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tqa">TQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.acl-short.123.mp4"/>
     </paper>
     <paper id="124">
       <title>Entity Concept-enhanced Few-shot Relation Extraction</title>
@@ -11052,9 +11059,9 @@
       <url hash="db788b1f">2021.acl-short.124</url>
       <doi>10.18653/v1/2021.acl-short.124</doi>
       <bibkey>yang-etal-2021-entity</bibkey>
+      <video href="2021.acl-short.124.mp4"/>
       <pwccode url="https://github.com/LittleGuoKe/ConceptFERE" additional="false">LittleGuoKe/ConceptFERE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
-      <video href="2021.acl-short.124.mp4"/>
     </paper>
     <paper id="125">
       <title>Improving Model Generalization: A <fixed-case>C</fixed-case>hinese Named Entity Recognition Case Study</title>
@@ -11080,8 +11087,8 @@
       <url hash="40942c47">2021.acl-short.126</url>
       <doi>10.18653/v1/2021.acl-short.126</doi>
       <bibkey>huang-etal-2021-three</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/docred">DocRED</pwcdataset>
       <video href="2021.acl-short.126.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/docred">DocRED</pwcdataset>
     </paper>
     <paper id="127">
       <title>Unsupervised Cross-Domain Prerequisite Chain Learning using Variational Graph Autoencoders</title>
@@ -11096,8 +11103,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="b248c6f4">2021.acl-short.127.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.127</doi>
       <bibkey>li-etal-2021-unsupervised-cross</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/lecturebank">LectureBank</pwcdataset>
       <video href="2021.acl-short.127.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/lecturebank">LectureBank</pwcdataset>
     </paper>
     <paper id="128">
       <title>Attentive Multiview Text Representation for Differential Diagnosis</title>
@@ -11122,11 +11129,11 @@
       <attachment type="OptionalSupplementaryMaterial" hash="23ab0a62">2021.acl-short.129.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-short.129</doi>
       <bibkey>herlihy-rudinger-2021-mednli</bibkey>
+      <video href="2021.acl-short.129.mp4"/>
       <pwccode url="https://github.com/crherlihy/clinical_nli_artifacts" additional="false">crherlihy/clinical_nli_artifacts</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mimic-iii">MIMIC-III</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.acl-short.129.mp4"/>
     </paper>
     <paper id="130">
       <title>Towards a more Robust Evaluation for Conversational Question Answering</title>
@@ -11138,9 +11145,9 @@
       <url hash="040d1e09">2021.acl-short.130</url>
       <doi>10.18653/v1/2021.acl-short.130</doi>
       <bibkey>siblini-etal-2021-towards</bibkey>
+      <video href="2021.acl-short.130.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coqa">CoQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-short.130.mp4"/>
     </paper>
     <paper id="131">
       <title><fixed-case>VAULT</fixed-case>: <fixed-case>VA</fixed-case>riable Unified Long Text Representation for Machine Reading Comprehension</title>
@@ -11154,10 +11161,10 @@
       <url hash="a88ae210">2021.acl-short.131</url>
       <doi>10.18653/v1/2021.acl-short.131</doi>
       <bibkey>wen-etal-2021-vault</bibkey>
+      <video href="2021.acl-short.131.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/techqa">TechQA</pwcdataset>
-      <video href="2021.acl-short.131.mp4"/>
     </paper>
     <paper id="132">
       <title>Avoiding Overlap in Data Augmentation for <fixed-case>AMR</fixed-case>-to-Text Generation</title>
@@ -11180,8 +11187,8 @@
       <url hash="3a9162b9">2021.acl-short.133</url>
       <doi>10.18653/v1/2021.acl-short.133</doi>
       <bibkey>yang-etal-2021-weakly</bibkey>
-      <pwccode url="https://github.com/yangalan123/WM-SRA" additional="false">yangalan123/WM-SRA</pwccode>
       <video href="2021.acl-short.133.mp4"/>
+      <pwccode url="https://github.com/yangalan123/WM-SRA" additional="false">yangalan123/WM-SRA</pwccode>
     </paper>
     <paper id="134">
       <title>Can Transformer Models Measure Coherence In Text: Re-Thinking the Shuffle Test</title>
@@ -11194,8 +11201,8 @@
       <url hash="10abeef7">2021.acl-short.134</url>
       <doi>10.18653/v1/2021.acl-short.134</doi>
       <bibkey>laban-etal-2021-transformer</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/billsum">BillSum</pwcdataset>
       <video href="2021.acl-short.134.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/billsum">BillSum</pwcdataset>
     </paper>
     <paper id="135">
       <title><fixed-case>S</fixed-case>im<fixed-case>CLS</fixed-case>: A Simple Framework for Contrastive Learning of Abstractive Summarization</title>
@@ -11206,9 +11213,9 @@
       <url hash="0bbb883c">2021.acl-short.135</url>
       <doi>10.18653/v1/2021.acl-short.135</doi>
       <bibkey>liu-liu-2021-simcls</bibkey>
+      <video href="2021.acl-short.135.mp4"/>
       <pwccode url="https://github.com/yixinL7/SimCLS" additional="true">yixinL7/SimCLS</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
-      <video href="2021.acl-short.135.mp4"/>
     </paper>
     <paper id="136">
       <title><fixed-case>S</fixed-case>a<fixed-case>R</fixed-case>o<fixed-case>C</fixed-case>o: Detecting Satire in a Novel <fixed-case>R</fixed-case>omanian Corpus of News Articles</title>
@@ -11220,9 +11227,9 @@
       <url hash="d8416ee4">2021.acl-short.136</url>
       <doi>10.18653/v1/2021.acl-short.136</doi>
       <bibkey>rogoz-etal-2021-saroco</bibkey>
+      <video href="2021.acl-short.136.mp4"/>
       <pwccode url="https://github.com/MihaelaGaman/SaRoCo" additional="false">MihaelaGaman/SaRoCo</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/saroco">SaRoCo</pwcdataset>
-      <video href="2021.acl-short.136.mp4"/>
     </paper>
     <paper id="137">
       <title>Bringing Structure into Summaries: a Faceted Summarization Dataset for Long Scientific Documents</title>
@@ -11238,9 +11245,9 @@
       <url hash="b2be586b">2021.acl-short.137</url>
       <doi>10.18653/v1/2021.acl-short.137</doi>
       <bibkey>meng-etal-2021-bringing</bibkey>
+      <video href="2021.acl-short.137.mp4"/>
       <pwccode url="https://github.com/hfthair/emerald_crawler" additional="false">hfthair/emerald_crawler</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/facetsum">FacetSum</pwcdataset>
-      <video href="2021.acl-short.137.mp4"/>
     </paper>
     <paper id="138">
       <title><fixed-case>R</fixed-case>eplicating and Extending “<fixed-case>B</fixed-case>ecause Their Treebanks Leak”: <fixed-case>G</fixed-case>raph Isomorphism, Covariants, and Parser Performance</title>
@@ -11312,9 +11319,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="a6707315">2021.acl-srw.2.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-srw.2</doi>
       <bibkey>wang-etal-2021-stage</bibkey>
+      <video href="2021.acl-srw.2.mp4"/>
       <pwccode url="https://github.com/EagleW/Stage-wise-Fine-tuning" additional="false">EagleW/Stage-wise-Fine-tuning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.acl-srw.2.mp4"/>
     </paper>
     <paper id="3">
       <title>Transformer-Based Direct Hidden <fixed-case>M</fixed-case>arkov Model for Machine Translation</title>
@@ -11353,8 +11360,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="957d5191">2021.acl-srw.5.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-srw.5</doi>
       <bibkey>bansal-etal-2021-low</bibkey>
-      <pwccode url="https://github.com/cdli-gh/Semi-Supervised-NMT-for-Sumerian-English" additional="true">cdli-gh/Semi-Supervised-NMT-for-Sumerian-English</pwccode>
       <video href="2021.acl-srw.5.mp4"/>
+      <pwccode url="https://github.com/cdli-gh/Semi-Supervised-NMT-for-Sumerian-English" additional="true">cdli-gh/Semi-Supervised-NMT-for-Sumerian-English</pwccode>
     </paper>
     <paper id="6">
       <title>On the Relationship between <fixed-case>Z</fixed-case>ipf’s Law of Abbreviation and Interfering Noise in Emergent Languages</title>
@@ -11385,9 +11392,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="a29f0c1d">2021.acl-srw.7.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-srw.7</doi>
       <bibkey>bajaj-etal-2021-long</bibkey>
+      <video href="2021.acl-srw.7.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
-      <video href="2021.acl-srw.7.mp4"/>
     </paper>
     <paper id="8">
       <title>Attending Self-Attention: A Case Study of Visually Grounded Supervision in Vision-and-Language Transformers</title>
@@ -11402,8 +11409,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="f95f6ac2">2021.acl-srw.8.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.acl-srw.8</doi>
       <bibkey>samaran-etal-2021-attending</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <video href="2021.acl-srw.8.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
     </paper>
     <paper id="9">
       <title>Video-guided Machine Translation with Spatial Hierarchical Attention Network</title>
@@ -11436,9 +11443,9 @@
       <url hash="5b8d534e">2021.acl-srw.11</url>
       <doi>10.18653/v1/2021.acl-srw.11</doi>
       <bibkey>testoni-bernardi-2021-ive</bibkey>
+      <video href="2021.acl-srw.11.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/guesswhat">GuessWhat?!</pwcdataset>
-      <video href="2021.acl-srw.11.mp4"/>
     </paper>
     <paper id="12">
       <title>How do different factors Impact the Inter-language Similarity? A Case Study on <fixed-case>I</fixed-case>ndian languages</title>
@@ -11476,8 +11483,8 @@
       <url hash="24f19583">2021.acl-srw.14</url>
       <doi>10.18653/v1/2021.acl-srw.14</doi>
       <bibkey>smirnova-etal-2021-situation</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/tweebank">Tweebank</pwcdataset>
       <video href="2021.acl-srw.14.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/tweebank">Tweebank</pwcdataset>
     </paper>
     <paper id="15">
       <title>Modeling Text using the Continuous Space Topic Model with Pre-Trained Word Embeddings</title>
@@ -11507,9 +11514,9 @@
       <attachment type="OptionalSupplementaryMaterial" hash="8b4241eb">2021.acl-srw.16.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-srw.16</doi>
       <bibkey>bai-etal-2021-semantics</bibkey>
+      <video href="2021.acl-srw.16.mp4"/>
       <pwccode url="https://github.com/rsvp-ai/semantic_unwritten" additional="false">rsvp-ai/semantic_unwritten</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
-      <video href="2021.acl-srw.16.mp4"/>
     </paper>
     <paper id="17">
       <title>Data Augmentation with Unsupervised Machine Translation Improves the Structural Similarity of Cross-lingual Word Embeddings</title>
@@ -11521,9 +11528,9 @@
       <url hash="a9c0107a">2021.acl-srw.17</url>
       <doi>10.18653/v1/2021.acl-srw.17</doi>
       <bibkey>nishikawa-etal-2021-data</bibkey>
+      <video href="2021.acl-srw.17.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/mldoc">MLDoc</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.acl-srw.17.mp4"/>
     </paper>
     <paper id="18">
       <title>Joint Detection and Coreference Resolution of Entities and Events with Document-level Context Aggregation</title>
@@ -11571,11 +11578,11 @@
       <url hash="337ea3b8">2021.acl-srw.21</url>
       <doi>10.18653/v1/2021.acl-srw.21</doi>
       <bibkey>shinoda-etal-2021-improving</bibkey>
+      <video href="2021.acl-srw.21.mp4"/>
       <pwccode url="https://github.com/KazutoshiShinoda/VQAG" additional="false">KazutoshiShinoda/VQAG</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsqa">NewsQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-srw.21.mp4"/>
     </paper>
     <paper id="22">
       <title>Tools Impact on the Quality of Annotations for Chat Untangling</title>
@@ -11623,8 +11630,8 @@
       <attachment type="OptionalSupplementaryMaterial" hash="061c0a85">2021.acl-srw.25.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-srw.25</doi>
       <bibkey>ficsor-berend-2021-changing</bibkey>
-      <pwccode url="https://github.com/ficstamas/word_embedding_interpretability" additional="false">ficstamas/word_embedding_interpretability</pwccode>
       <video href="2021.acl-srw.25.mp4"/>
+      <pwccode url="https://github.com/ficstamas/word_embedding_interpretability" additional="false">ficstamas/word_embedding_interpretability</pwccode>
     </paper>
     <paper id="26">
       <title>Personal Bias in Prediction of Emotions Elicited by Textual Opinions</title>
@@ -11649,10 +11656,10 @@
       <url hash="1b68ccf9">2021.acl-srw.27</url>
       <doi>10.18653/v1/2021.acl-srw.27</doi>
       <bibkey>zhu-2021-mvp</bibkey>
+      <video href="2021.acl-srw.27.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cmrc">CMRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cmrc-2018">CMRC 2018</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/chid">ChID</pwcdataset>
-      <video href="2021.acl-srw.27.mp4"/>
     </paper>
     <paper id="28">
       <title><fixed-case>CMTA</fixed-case>: <fixed-case>COVID</fixed-case>-19 Misinformation Multilingual Analysis on <fixed-case>T</fixed-case>witter</title>
@@ -11692,9 +11699,9 @@
       <url hash="46b12c0b">2021.acl-srw.30</url>
       <doi>10.18653/v1/2021.acl-srw.30</doi>
       <bibkey>gupta-etal-2021-sumpubmed</bibkey>
+      <video href="2021.acl-srw.30.mp4"/>
       <pwccode url="https://github.com/vgupta123/sumpubmed" additional="false">vgupta123/sumpubmed</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
-      <video href="2021.acl-srw.30.mp4"/>
     </paper>
     <paper id="31">
       <title>A Case Study of Analysis of Construals in Language on Social Media Surrounding a Crisis Event</title>
@@ -11719,8 +11726,8 @@
       <url hash="b28576c3">2021.acl-srw.32</url>
       <doi>10.18653/v1/2021.acl-srw.32</doi>
       <bibkey>dementieva-panchenko-2021-cross</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/nela-gt-2018">NELA-GT-2018</pwcdataset>
       <video href="2021.acl-srw.32.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/nela-gt-2018">NELA-GT-2018</pwcdataset>
     </paper>
     <paper id="33">
       <title>Neural Machine Translation with Synchronous Latent Phrase Structure</title>
@@ -11731,8 +11738,8 @@
       <url hash="ddfd8c9d">2021.acl-srw.33</url>
       <doi>10.18653/v1/2021.acl-srw.33</doi>
       <bibkey>harada-watanabe-2021-neural</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/aspec">ASPEC</pwcdataset>
       <video href="2021.acl-srw.33.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/aspec">ASPEC</pwcdataset>
     </paper>
     <paper id="34">
       <title>Zero Pronouns Identification based on Span prediction</title>
@@ -11758,8 +11765,8 @@
       <url hash="75238265">2021.acl-srw.35</url>
       <doi>10.18653/v1/2021.acl-srw.35</doi>
       <bibkey>vazquez-etal-2021-differences</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
       <video href="2021.acl-srw.35.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
     </paper>
     <paper id="36">
       <title>Synchronous Syntactic Attention for Transformer Neural Machine Translation</title>
@@ -11826,8 +11833,8 @@
       <url hash="e56bd3d6">2021.acl-demo.2</url>
       <doi>10.18653/v1/2021.acl-demo.2</doi>
       <bibkey>lee-etal-2021-intellicat</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wmt-2020">WMT 2020</pwcdataset>
       <video href="2021.acl-demo.2.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wmt-2020">WMT 2020</pwcdataset>
     </paper>
     <paper id="3">
       <title>The <fixed-case>C</fixed-case>lassical <fixed-case>L</fixed-case>anguage <fixed-case>T</fixed-case>oolkit: <fixed-case>A</fixed-case>n <fixed-case>NLP</fixed-case> Framework for Pre-Modern Languages</title>
@@ -11842,8 +11849,8 @@
       <url hash="e10171d9">2021.acl-demo.3</url>
       <doi>10.18653/v1/2021.acl-demo.3</doi>
       <bibkey>johnson-etal-2021-classical</bibkey>
-      <pwccode url="https://github.com/cltk/cltk" additional="false">cltk/cltk</pwccode>
       <video href="2021.acl-demo.3.mp4"/>
+      <pwccode url="https://github.com/cltk/cltk" additional="false">cltk/cltk</pwccode>
     </paper>
     <paper id="4">
       <title><fixed-case>T</fixed-case>ext<fixed-case>B</fixed-case>ox: A Unified, Modularized, and Extensible Framework for Text Generation</title>
@@ -11862,10 +11869,10 @@
       <url hash="dbe24eb2">2021.acl-demo.4</url>
       <doi>10.18653/v1/2021.acl-demo.4</doi>
       <bibkey>li-etal-2021-textbox</bibkey>
+      <video href="2021.acl-demo.4.mp4"/>
       <pwccode url="https://github.com/RUCAIBox/TextBox" additional="false">RUCAIBox/TextBox</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.acl-demo.4.mp4"/>
     </paper>
     <paper id="5">
       <title>Inside <fixed-case>ASCENT</fixed-case>: Exploring a Deep Commonsense Knowledge Base and its Usage in Question Answering</title>
@@ -11877,12 +11884,12 @@
       <url hash="0220d20d">2021.acl-demo.5</url>
       <doi>10.18653/v1/2021.acl-demo.5</doi>
       <bibkey>nguyen-etal-2021-inside</bibkey>
+      <video href="2021.acl-demo.5.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/atomic">ATOMIC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ascentkb">Ascent KB</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quasimodo">Quasimodo</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webchild">WebChild</pwcdataset>
-      <video href="2021.acl-demo.5.mp4"/>
     </paper>
     <paper id="6">
       <title><fixed-case>S</fixed-case>ci<fixed-case>C</fixed-case>oncept<fixed-case>M</fixed-case>iner: A system for large-scale scientific concept discovery</title>
@@ -11896,8 +11903,8 @@
       <url hash="3e4043b7">2021.acl-demo.6</url>
       <doi>10.18653/v1/2021.acl-demo.6</doi>
       <bibkey>shen-etal-2021-sciconceptminer</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/microsoft-academic-graph">Microsoft Academic Graph</pwcdataset>
       <video href="2021.acl-demo.6.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/microsoft-academic-graph">Microsoft Academic Graph</pwcdataset>
     </paper>
     <paper id="7">
       <title><fixed-case>N</fixed-case>eur<fixed-case>ST</fixed-case>: Neural Speech Translation Toolkit</title>
@@ -11986,8 +11993,8 @@
       <url hash="d599a858">2021.acl-demo.12</url>
       <doi>10.18653/v1/2021.acl-demo.12</doi>
       <bibkey>geng-etal-2021-fasthan</bibkey>
-      <pwccode url="https://github.com/fastnlp/fastHan" additional="false">fastnlp/fastHan</pwccode>
       <video href="2021.acl-demo.12.mp4"/>
+      <pwccode url="https://github.com/fastnlp/fastHan" additional="false">fastnlp/fastHan</pwccode>
     </paper>
     <paper id="13">
       <title>Erase and Rewind: Manual Correction of <fixed-case>NLP</fixed-case> Output through a Web Interface</title>
@@ -12017,9 +12024,9 @@
       <award>Best Demonstration Runner-up</award>
       <doi>10.18653/v1/2021.acl-demo.14</doi>
       <bibkey>hongwimol-etal-2021-esra</bibkey>
+      <video href="2021.acl-demo.14.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semantic-scholar">Semantic Scholar</pwcdataset>
-      <video href="2021.acl-demo.14.mp4"/>
     </paper>
     <paper id="15">
       <title>Trafilatura: <fixed-case>A</fixed-case> Web Scraping Library and Command-Line Tool for Text Discovery and Extraction</title>
@@ -12029,8 +12036,8 @@
       <url hash="5fc847d1">2021.acl-demo.15</url>
       <doi>10.18653/v1/2021.acl-demo.15</doi>
       <bibkey>barbaresi-2021-trafilatura</bibkey>
-      <pwccode url="https://github.com/adbar/trafilatura" additional="false">adbar/trafilatura</pwccode>
       <video href="2021.acl-demo.15.mp4"/>
+      <pwccode url="https://github.com/adbar/trafilatura" additional="false">adbar/trafilatura</pwccode>
     </paper>
     <paper id="16">
       <title>Dodrio: Exploring Transformer Models with Interactive Visualization</title>
@@ -12043,10 +12050,10 @@
       <attachment type="OptionalSupplementaryMaterial" hash="65ea0e5b">2021.acl-demo.16.OptionalSupplementaryMaterial.zip</attachment>
       <doi>10.18653/v1/2021.acl-demo.16</doi>
       <bibkey>wang-etal-2021-dodrio</bibkey>
+      <video href="2021.acl-demo.16.mp4"/>
       <pwccode url="https://github.com/poloclub/dodrio" additional="false">poloclub/dodrio</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.acl-demo.16.mp4"/>
     </paper>
     <paper id="17">
       <title><fixed-case>REM</fixed-case>: Efficient Semi-Automated Real-Time Moderation of Online Forums</title>
@@ -12071,9 +12078,9 @@
       <url hash="410f5360">2021.acl-demo.18</url>
       <doi>10.18653/v1/2021.acl-demo.18</doi>
       <bibkey>vig-etal-2021-summvis</bibkey>
+      <video href="2021.acl-demo.18.mp4"/>
       <pwccode url="https://github.com/robustness-gym/summvis" additional="false">robustness-gym/summvis</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/newsroom">NEWSROOM</pwcdataset>
-      <video href="2021.acl-demo.18.mp4"/>
     </paper>
     <paper id="19">
       <title>A Graphical Interface for Curating Schemas</title>
@@ -12102,9 +12109,9 @@
       <url hash="820068b5">2021.acl-demo.20</url>
       <doi>10.18653/v1/2021.acl-demo.20</doi>
       <bibkey>zhang-etal-2021-textoir</bibkey>
+      <video href="2021.acl-demo.20.mp4"/>
       <pwccode url="https://github.com/thuiar/textoir" additional="true">thuiar/textoir</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.acl-demo.20.mp4"/>
     </paper>
     <paper id="21">
       <title><fixed-case>K</fixed-case>ui<fixed-case>L</fixed-case>ei<fixed-case>X</fixed-case>i: a <fixed-case>C</fixed-case>hinese Open-Ended Text Adventure Game</title>
@@ -12145,11 +12152,11 @@
       <url hash="6cc9f854">2021.acl-demo.22</url>
       <doi>10.18653/v1/2021.acl-demo.22</doi>
       <bibkey>zhou-etal-2021-crslab</bibkey>
+      <video href="2021.acl-demo.22.mp4"/>
       <pwccode url="https://github.com/RUCAIBox/CRSLab" additional="false">RUCAIBox/CRSLab</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/durecdial">DuRecDial</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/redial">ReDial</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tg-redial">TG-ReDial</pwcdataset>
-      <video href="2021.acl-demo.22.mp4"/>
     </paper>
     <paper id="23">
       <title>Does My Representation Capture <fixed-case>X</fixed-case>? Probe-Ably</title>
@@ -12177,9 +12184,9 @@
       <url hash="3b8d155d">2021.acl-demo.24</url>
       <doi>10.18653/v1/2021.acl-demo.24</doi>
       <bibkey>pan-etal-2021-cltr</bibkey>
+      <video href="2021.acl-demo.24.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitablequestions">WikiTableQuestions</pwcdataset>
-      <video href="2021.acl-demo.24.mp4"/>
     </paper>
     <paper id="25">
       <title>Neural Extractive Search</title>
@@ -12283,9 +12290,9 @@
       <url hash="e261d0b5">2021.acl-demo.30</url>
       <doi>10.18653/v1/2021.acl-demo.30</doi>
       <bibkey>alammar-2021-ecco</bibkey>
+      <video href="2021.acl-demo.30.mp4"/>
       <pwccode url="https://github.com/jalammar/ecco" additional="false">jalammar/ecco</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/11k-hands">11k Hands</pwcdataset>
-      <video href="2021.acl-demo.30.mp4"/>
     </paper>
     <paper id="31">
       <title><fixed-case>PAWLS</fixed-case>: <fixed-case>PDF</fixed-case> Annotation With Labels and Structure</title>
@@ -12297,10 +12304,10 @@
       <url hash="e867fda7">2021.acl-demo.31</url>
       <doi>10.18653/v1/2021.acl-demo.31</doi>
       <bibkey>neumann-etal-2021-pawls</bibkey>
+      <video href="2021.acl-demo.31.mp4"/>
       <pwccode url="https://github.com/allenai/pawls" additional="false">allenai/pawls</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/docbank">DocBank</pwcdataset>
-      <video href="2021.acl-demo.31.mp4"/>
     </paper>
     <paper id="32">
       <title><fixed-case>T</fixed-case>wee<fixed-case>NLP</fixed-case>: A <fixed-case>T</fixed-case>witter Exploration Portal for Natural Language Processing</title>
@@ -12324,8 +12331,8 @@
       <url hash="01fc5cdd">2021.acl-demo.33</url>
       <doi>10.18653/v1/2021.acl-demo.33</doi>
       <bibkey>zhang-etal-2021-chrentranslate</bibkey>
-      <pwccode url="https://github.com/ZhangShiyue/ChrEn" additional="true">ZhangShiyue/ChrEn</pwccode>
       <video href="2021.acl-demo.33.mp4"/>
+      <pwccode url="https://github.com/ZhangShiyue/ChrEn" additional="true">ZhangShiyue/ChrEn</pwccode>
     </paper>
     <paper id="34">
       <title><fixed-case>E</fixed-case>xplaina<fixed-case>B</fixed-case>oard: An Explainable Leaderboard for <fixed-case>NLP</fixed-case></title>
@@ -12344,9 +12351,9 @@
       <award>Best Demonstration Paper</award>
       <doi>10.18653/v1/2021.acl-demo.34</doi>
       <bibkey>liu-etal-2021-explainaboard</bibkey>
+      <video href="2021.acl-demo.34.mp4"/>
       <pwccode url="https://github.com/neulab/ExplainaBoard" additional="false">neulab/ExplainaBoard</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
-      <video href="2021.acl-demo.34.mp4"/>
     </paper>
     <paper id="35">
       <title>Exploring Word Usage Change with Continuously Evolving Embeddings</title>
@@ -12356,8 +12363,8 @@
       <url hash="6342ce85">2021.acl-demo.35</url>
       <doi>10.18653/v1/2021.acl-demo.35</doi>
       <bibkey>horn-2021-exploring</bibkey>
-      <pwccode url="https://github.com/cod3licious/evolvemb" additional="false">cod3licious/evolvemb</pwccode>
       <video href="2021.acl-demo.35.mp4"/>
+      <pwccode url="https://github.com/cod3licious/evolvemb" additional="false">cod3licious/evolvemb</pwccode>
     </paper>
     <paper id="36">
       <title><fixed-case>TURING</fixed-case>: an Accurate and Interpretable Multi-Hypothesis Cross-Domain Natural Language Database Interface</title>
@@ -12376,8 +12383,8 @@
       <url hash="586c232e">2021.acl-demo.36</url>
       <doi>10.18653/v1/2021.acl-demo.36</doi>
       <bibkey>xu-etal-2021-turing</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
       <video href="2021.acl-demo.36.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
     </paper>
     <paper id="37">
       <title>Many-to-<fixed-case>E</fixed-case>nglish Machine Translation Tools, Data, and Pretrained Models</title>
@@ -12390,9 +12397,9 @@
       <url hash="a1c03e0f">2021.acl-demo.37</url>
       <doi>10.18653/v1/2021.acl-demo.37</doi>
       <bibkey>gowda-etal-2021-many</bibkey>
+      <video href="2021.acl-demo.37.mp4"/>
       <pwccode url="https://github.com/thammegowda/006-many-to-eng" additional="true">thammegowda/006-many-to-eng</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/opus-100">OPUS-100</pwcdataset>
-      <video href="2021.acl-demo.37.mp4"/>
     </paper>
     <paper id="38">
       <title><fixed-case>LEGOE</fixed-case>val: An Open-Source Toolkit for Dialogue System Evaluation via Crowdsourcing</title>
@@ -12406,8 +12413,8 @@
       <url hash="9ed8bc3e">2021.acl-demo.38</url>
       <doi>10.18653/v1/2021.acl-demo.38</doi>
       <bibkey>li-etal-2021-legoeval</bibkey>
-      <pwccode url="https://github.com/yooli23/LEGOEval" additional="false">yooli23/LEGOEval</pwccode>
       <video href="2021.acl-demo.38.mp4"/>
+      <pwccode url="https://github.com/yooli23/LEGOEval" additional="false">yooli23/LEGOEval</pwccode>
     </paper>
     <paper id="39">
       <title><fixed-case>R</fixed-case>e<fixed-case>T</fixed-case>ra<fixed-case>C</fixed-case>k: A Flexible and Efficient Framework for Knowledge Base Question Answering</title>
@@ -12422,10 +12429,10 @@
       <url hash="e07c4f50">2021.acl-demo.39</url>
       <doi>10.18653/v1/2021.acl-demo.39</doi>
       <bibkey>chen-etal-2021-retrack</bibkey>
+      <video href="2021.acl-demo.39.mp4"/>
       <pwccode url="https://github.com/microsoft/KC/tree/main/papers/ReTraCk" additional="false">microsoft/KC</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/grailqa">GrailQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestionssp">WebQuestionsSP</pwcdataset>
-      <video href="2021.acl-demo.39.mp4"/>
     </paper>
     <paper id="40">
       <title>skweak: Weak Supervision Made Easy for <fixed-case>NLP</fixed-case></title>
@@ -12483,9 +12490,9 @@
       <url hash="52f58f1f">2021.acl-demo.41</url>
       <doi>10.18653/v1/2021.acl-demo.41</doi>
       <bibkey>wang-etal-2021-textflint</bibkey>
+      <video href="2021.acl-demo.41.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.acl-demo.41.mp4"/>
     </paper>
     <paper id="42">
       <title>Stretch-<fixed-case>VST</fixed-case>: Getting Flexible With Visual Stories</title>
@@ -12499,8 +12506,8 @@
       <url hash="525668fc">2021.acl-demo.42</url>
       <doi>10.18653/v1/2021.acl-demo.42</doi>
       <bibkey>hsu-etal-2021-stretch</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/vist">VIST</pwcdataset>
       <video href="2021.acl-demo.42.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/vist">VIST</pwcdataset>
     </paper>
     <paper id="43">
       <title><fixed-case>O</fixed-case>pen<fixed-case>A</fixed-case>ttack: An Open-source Textual Adversarial Attack Toolkit</title>
diff --git a/data/xml/2021.conll.xml b/data/xml/2021.conll.xml
index ba6728f2c8..f160cf9183 100644
--- a/data/xml/2021.conll.xml
+++ b/data/xml/2021.conll.xml
@@ -450,6 +450,7 @@
       <url hash="5b2e87c6">2021.conll-1.32</url>
       <bibkey>huebner-willits-2021-scaffolded</bibkey>
       <doi>10.18653/v1/2021.conll-1.32</doi>
+      <pwccode url="https://github.com/phueb/entropic" additional="false">phueb/entropic</pwccode>
     </paper>
     <paper id="33">
       <title>Grammatical Profiling for Semantic Change Detection</title>
diff --git a/data/xml/2021.emnlp.xml b/data/xml/2021.emnlp.xml
index 724daf10ff..a907640c3e 100644
--- a/data/xml/2021.emnlp.xml
+++ b/data/xml/2021.emnlp.xml
@@ -43,8 +43,8 @@
       <url hash="3327c149">2021.emnlp-main.2</url>
       <bibkey>chen-etal-2021-zero</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.2</doi>
-      <pwccode url="https://github.com/ghchen18/emnlp2021-sixt" additional="false">ghchen18/emnlp2021-sixt</pwccode>
       <video href="2021.emnlp-main.2.mp4"/>
+      <pwccode url="https://github.com/ghchen18/emnlp2021-sixt" additional="false">ghchen18/emnlp2021-sixt</pwccode>
     </paper>
     <paper id="3">
       <title><fixed-case>ERNIE</fixed-case>-<fixed-case>M</fixed-case>: Enhanced Multilingual Representation by Aligning Cross-lingual Semantics with Monolingual Corpora</title>
@@ -60,10 +60,10 @@
       <url hash="0837dc82">2021.emnlp-main.3</url>
       <bibkey>ouyang-etal-2021-ernie</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.3</doi>
+      <video href="2021.emnlp-main.3.mp4"/>
       <pwccode url="https://github.com/PaddlePaddle/ERNIE" additional="true">PaddlePaddle/ERNIE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xtreme">XTREME</pwcdataset>
-      <video href="2021.emnlp-main.3.mp4"/>
     </paper>
     <paper id="4">
       <title>Cross Attention Augmented Transducer Networks for Simultaneous Translation</title>
@@ -77,8 +77,8 @@
       <url hash="746f2138">2021.emnlp-main.4</url>
       <bibkey>liu-etal-2021-cross</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.4</doi>
-      <pwccode url="https://github.com/danliu2/caat" additional="false">danliu2/caat</pwccode>
       <video href="2021.emnlp-main.4.mp4"/>
+      <pwccode url="https://github.com/danliu2/caat" additional="false">danliu2/caat</pwccode>
     </paper>
     <paper id="5">
       <title>Translating Headers of Tabular Data: A Pilot Study of Schema Translation</title>
@@ -91,8 +91,8 @@
       <url hash="401bcc32">2021.emnlp-main.5</url>
       <bibkey>zhu-etal-2021-translating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.5</doi>
-      <pwccode url="https://github.com/microsoft/ContextualSP" additional="false">microsoft/ContextualSP</pwccode>
       <video href="2021.emnlp-main.5.mp4"/>
+      <pwccode url="https://github.com/microsoft/ContextualSP" additional="false">microsoft/ContextualSP</pwccode>
     </paper>
     <paper id="6">
       <title>Towards Making the Most of Dialogue Characteristics for Neural Chat Translation</title>
@@ -108,9 +108,9 @@
       <url hash="c9f3412a">2021.emnlp-main.6</url>
       <bibkey>liang-etal-2021-towards</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.6</doi>
+      <video href="2021.emnlp-main.6.mp4"/>
       <pwccode url="https://github.com/xl2248/csa-nct" additional="false">xl2248/csa-nct</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bmeld">BMELD</pwcdataset>
-      <video href="2021.emnlp-main.6.mp4"/>
     </paper>
     <paper id="7">
       <title>Low-Resource Dialogue Summarization with Domain-Agnostic Multi-Source Pretraining</title>
@@ -124,12 +124,12 @@
       <url hash="c734b477">2021.emnlp-main.7</url>
       <bibkey>zou-etal-2021-low</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.7</doi>
+      <video href="2021.emnlp-main.7.mp4"/>
       <pwccode url="https://github.com/rowitzou/dams" additional="false">rowitzou/dams</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/reddit-conversation-corpus">Reddit Conversation Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.emnlp-main.7.mp4"/>
     </paper>
     <paper id="8">
       <title>Controllable Neural Dialogue Summarization with Personal Named Entity Planning</title>
@@ -140,8 +140,8 @@
       <url hash="8adb9328">2021.emnlp-main.8</url>
       <bibkey>liu-chen-2021-controllable</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.8</doi>
-      <pwccode url="https://github.com/seq-to-mind/planning_dial_summ" additional="false">seq-to-mind/planning_dial_summ</pwccode>
       <video href="2021.emnlp-main.8.mp4"/>
+      <pwccode url="https://github.com/seq-to-mind/planning_dial_summ" additional="false">seq-to-mind/planning_dial_summ</pwccode>
     </paper>
     <paper id="9">
       <title>Fine-grained Factual Consistency Assessment for Abstractive Summarization Models</title>
@@ -153,8 +153,8 @@
       <url hash="95b1e684">2021.emnlp-main.9</url>
       <bibkey>zhang-etal-2021-fine-grained</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.9</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <video href="2021.emnlp-main.9.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
     </paper>
     <paper id="10">
       <title>Decision-Focused Summarization</title>
@@ -165,8 +165,8 @@
       <url hash="77f04c29">2021.emnlp-main.10</url>
       <bibkey>hsu-tan-2021-decision</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.10</doi>
-      <pwccode url="https://github.com/chicagohai/decsum" additional="false">chicagohai/decsum</pwccode>
       <video href="2021.emnlp-main.10.mp4"/>
+      <pwccode url="https://github.com/chicagohai/decsum" additional="false">chicagohai/decsum</pwccode>
     </paper>
     <paper id="11">
       <title>Multiplex Graph Neural Network for Extractive Text Summarization</title>
@@ -208,8 +208,8 @@
       <attachment type="Software" hash="41b48d72">2021.emnlp-main.13.Software.zip</attachment>
       <bibkey>liu-etal-2021-hetformer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.13</doi>
-      <pwccode url="https://github.com/yeliu918/hetformer" additional="false">yeliu918/hetformer</pwccode>
       <video href="2021.emnlp-main.13.mp4"/>
+      <pwccode url="https://github.com/yeliu918/hetformer" additional="false">yeliu918/hetformer</pwccode>
     </paper>
     <paper id="14">
       <title>Unsupervised Keyphrase Extraction by Jointly Modeling Local and Global Context</title>
@@ -222,8 +222,8 @@
       <url hash="ebe6c57b">2021.emnlp-main.14</url>
       <bibkey>liang-etal-2021-unsupervised</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.14</doi>
-      <pwccode url="https://github.com/xnliang98/uke_ccrank" additional="false">xnliang98/uke_ccrank</pwccode>
       <video href="2021.emnlp-main.14.mp4"/>
+      <pwccode url="https://github.com/xnliang98/uke_ccrank" additional="false">xnliang98/uke_ccrank</pwccode>
     </paper>
     <paper id="15">
       <title>Distantly Supervised Relation Extraction using Multi-Layer Revision Network and Confidence-based Multi-Instance Learning</title>
@@ -249,10 +249,10 @@
       <url hash="978f6b9a">2021.emnlp-main.16</url>
       <bibkey>shi-etal-2021-logic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.16</doi>
+      <video href="2021.emnlp-main.16.mp4"/>
       <pwccode url="https://github.com/qshi95/lergv" additional="false">qshi95/lergv</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tabfact">TabFact</pwcdataset>
-      <video href="2021.emnlp-main.16.mp4"/>
     </paper>
     <paper id="17">
       <title>A Partition Filter Network for Joint Entity and Relation Extraction</title>
@@ -266,12 +266,12 @@
       <url hash="5183a8fd">2021.emnlp-main.17</url>
       <bibkey>yan-etal-2021-partition</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.17</doi>
+      <video href="2021.emnlp-main.17.mp4"/>
       <pwccode url="https://github.com/Coopercoppers/PFN" additional="false">Coopercoppers/PFN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2004">ACE 2004</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2005">ACE 2005</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.emnlp-main.17.mp4"/>
     </paper>
     <paper id="18">
       <title><fixed-case>TEBNER</fixed-case>: Domain Specific Named Entity Recognition with Type Expanded Boundary-aware Network</title>
@@ -287,9 +287,9 @@
       <url hash="a51fa8fe">2021.emnlp-main.18</url>
       <bibkey>fang-etal-2021-tebner</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.18</doi>
+      <video href="2021.emnlp-main.18.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bc5cdr">BC5CDR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ncbi-disease-1">NCBI Disease</pwcdataset>
-      <video href="2021.emnlp-main.18.mp4"/>
     </paper>
     <paper id="19">
       <title>Beta Distribution Guided Aspect-aware Graph for Aspect Category Sentiment Analysis with Affective Knowledge</title>
@@ -307,9 +307,9 @@
       <attachment type="Software" hash="48239545">2021.emnlp-main.19.Software.zip</attachment>
       <bibkey>liang-etal-2021-beta</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.19</doi>
+      <video href="2021.emnlp-main.19.mp4"/>
       <pwccode url="https://github.com/binliang-nlp/aagcn-acsa" additional="false">binliang-nlp/aagcn-acsa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
-      <video href="2021.emnlp-main.19.mp4"/>
     </paper>
     <paper id="20">
       <title><fixed-case>DILBERT</fixed-case>: Customized Pre-Training for Domain Adaptation with Category Shift, with an Application to Aspect Extraction</title>
@@ -321,9 +321,9 @@
       <url hash="be594994">2021.emnlp-main.20</url>
       <bibkey>lekhtman-etal-2021-dilbert</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.20</doi>
+      <video href="2021.emnlp-main.20.mp4"/>
       <pwccode url="https://github.com/tonylekhtman/dilbert" additional="false">tonylekhtman/dilbert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mams">MAMS</pwcdataset>
-      <video href="2021.emnlp-main.20.mp4"/>
     </paper>
     <paper id="21">
       <title>Improving Multimodal fusion via Mutual Dependency Maximisation</title>
@@ -336,8 +336,8 @@
       <url hash="939c7b0b">2021.emnlp-main.21</url>
       <bibkey>colombo-etal-2021-improving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.21</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/cmu-mosei">CMU-MOSEI</pwcdataset>
       <video href="2021.emnlp-main.21.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cmu-mosei">CMU-MOSEI</pwcdataset>
     </paper>
     <paper id="22">
       <title>Learning Implicit Sentiment in Aspect-based Sentiment Analysis with Supervised Contrastive Pre-Training</title>
@@ -351,9 +351,9 @@
       <url hash="9f0f5e71">2021.emnlp-main.22</url>
       <bibkey>li-etal-2021-learning-implicit</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.22</doi>
+      <video href="2021.emnlp-main.22.mp4"/>
       <pwccode url="https://github.com/tribleave/scapt-absa" additional="false">tribleave/scapt-absa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mams">MAMS</pwcdataset>
-      <video href="2021.emnlp-main.22.mp4"/>
     </paper>
     <paper id="23">
       <title>Progressive Self-Training with Discriminator for Aspect Term Extraction</title>
@@ -379,9 +379,9 @@
       <url hash="83269a66">2021.emnlp-main.24</url>
       <bibkey>chen-etal-2021-reinforced</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.24</doi>
+      <video href="2021.emnlp-main.24.mp4"/>
       <pwccode url="https://github.com/nustm/rcda" additional="false">nustm/rcda</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.24.mp4"/>
     </paper>
     <paper id="25">
       <title>Idiosyncratic but not Arbitrary: Learning Idiolects in Online Registers Reveals Distinctive yet Consistent Individual Styles</title>
@@ -392,8 +392,8 @@
       <url hash="5bfd7e81">2021.emnlp-main.25</url>
       <bibkey>zhu-jurgens-2021-idiosyncratic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.25</doi>
-      <pwccode url="https://github.com/lingjzhu/idiolect" additional="false">lingjzhu/idiolect</pwccode>
       <video href="2021.emnlp-main.25.mp4"/>
+      <pwccode url="https://github.com/lingjzhu/idiolect" additional="false">lingjzhu/idiolect</pwccode>
     </paper>
     <paper id="26">
       <title>Narrative Theory for Computational Narrative Understanding</title>
@@ -436,8 +436,8 @@
       <url hash="a81e529a">2021.emnlp-main.28</url>
       <bibkey>sen-etal-2021-counterfactually</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.28</doi>
-      <pwccode url="https://github.com/gesiscss/socialcad" additional="false">gesiscss/socialcad</pwccode>
       <video href="2021.emnlp-main.28.mp4"/>
+      <pwccode url="https://github.com/gesiscss/socialcad" additional="false">gesiscss/socialcad</pwccode>
     </paper>
     <paper id="29">
       <title>Latent Hatred: A Benchmark for Understanding Implicit Hate Speech</title>
@@ -453,12 +453,12 @@
       <url hash="badbed00">2021.emnlp-main.29</url>
       <bibkey>elsherief-etal-2021-latent</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.29</doi>
+      <video href="2021.emnlp-main.29.mp4"/>
       <pwccode url="https://github.com/gt-salt/implicit-hate" additional="false">gt-salt/implicit-hate</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/implicit-hate">Implicit Hate</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hate-speech">Hate Speech</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.emnlp-main.29.mp4"/>
     </paper>
     <paper id="30">
       <title>Distilling Linguistic Context for Language Model Compression</title>
@@ -470,11 +470,11 @@
       <url hash="af511c21">2021.emnlp-main.30</url>
       <bibkey>park-etal-2021-distilling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.30</doi>
+      <video href="2021.emnlp-main.30.mp4"/>
       <pwccode url="https://github.com/geondopark/ckd" additional="false">geondopark/ckd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.30.mp4"/>
     </paper>
     <paper id="31">
       <title>Dynamic Knowledge Distillation for Pre-trained Language Models</title>
@@ -489,12 +489,12 @@
       <url hash="d3f2ade9">2021.emnlp-main.31</url>
       <bibkey>li-etal-2021-dynamic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.31</doi>
+      <video href="2021.emnlp-main.31.mp4"/>
       <pwccode url="https://github.com/lancopku/dynamickd" additional="false">lancopku/dynamickd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.31.mp4"/>
     </paper>
     <paper id="32">
       <title>Few-Shot Text Generation with Natural Language Instructions</title>
@@ -505,9 +505,9 @@
       <url hash="861ef6b4">2021.emnlp-main.32</url>
       <bibkey>schick-schutze-2021-shot</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.32</doi>
+      <video href="2021.emnlp-main.32.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/aeslc">AESLC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/reddit-tifu">Reddit TIFU</pwcdataset>
-      <video href="2021.emnlp-main.32.mp4"/>
     </paper>
     <paper id="33">
       <title><fixed-case>SOM</fixed-case>-<fixed-case>NCSCM</fixed-case> : An Efficient Neural <fixed-case>C</fixed-case>hinese Sentence Compression Model Enhanced with Self-Organizing Map</title>
@@ -536,9 +536,9 @@
       <url hash="e8ace5a1">2021.emnlp-main.34</url>
       <bibkey>kung-etal-2021-efficient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.34</doi>
+      <video href="2021.emnlp-main.34.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.emnlp-main.34.mp4"/>
     </paper>
     <paper id="35">
       <title><fixed-case>GOLD</fixed-case>: Improving Out-of-Scope Detection in Dialogues using Data Augmentation</title>
@@ -549,11 +549,11 @@
       <url hash="94658ebb">2021.emnlp-main.35</url>
       <bibkey>chen-yu-2021-gold</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.35</doi>
+      <video href="2021.emnlp-main.35.mp4"/>
       <pwccode url="https://github.com/asappresearch/gold" additional="false">asappresearch/gold</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rostd">ROSTD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/star">STAR</pwcdataset>
-      <video href="2021.emnlp-main.35.mp4"/>
     </paper>
     <paper id="36">
       <title>Graph Based Network with Contextualized Representations of Turns in Dialogue</title>
@@ -564,12 +564,12 @@
       <url hash="c037cdbb">2021.emnlp-main.36</url>
       <bibkey>lee-choi-2021-graph</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.36</doi>
+      <video href="2021.emnlp-main.36.mp4"/>
       <pwccode url="https://github.com/blacknoodle/tucore-gcn" additional="false">blacknoodle/tucore-gcn</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dialogre">DialogRE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/emorynlp">EmoryNLP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
-      <video href="2021.emnlp-main.36.mp4"/>
     </paper>
     <paper id="37">
       <title>Automatically Exposing Problems with Neural Dialog Models</title>
@@ -580,8 +580,8 @@
       <url hash="fb2b6f5c">2021.emnlp-main.37</url>
       <bibkey>yu-sagae-2021-automatically</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.37</doi>
-      <pwccode url="https://github.com/diandyu/trigger" additional="false">diandyu/trigger</pwccode>
       <video href="2021.emnlp-main.37.mp4"/>
+      <pwccode url="https://github.com/diandyu/trigger" additional="false">diandyu/trigger</pwccode>
     </paper>
     <paper id="38">
       <title><fixed-case>E</fixed-case>vent Coreference Data (Almost) for Free: <fixed-case>M</fixed-case>ining Hyperlinks from Online News</title>
@@ -592,9 +592,9 @@
       <url hash="347b18cc">2021.emnlp-main.38</url>
       <bibkey>bugert-gurevych-2021-event</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.38</doi>
+      <video href="2021.emnlp-main.38.mp4"/>
       <pwccode url="https://github.com/ukplab/emnlp2021-hypercoref-cdcr" additional="false">ukplab/emnlp2021-hypercoref-cdcr</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ecb">ECB+</pwcdataset>
-      <video href="2021.emnlp-main.38.mp4"/>
     </paper>
     <paper id="39">
       <title>Inducing Stereotypical Character Roles from Plot Structure</title>
@@ -631,10 +631,10 @@
       <attachment type="Software" hash="b79e0e4c">2021.emnlp-main.41.Software.zip</attachment>
       <bibkey>wolfe-caliskan-2021-low</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.41</doi>
+      <video href="2021.emnlp-main.41.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/openwebtext">OpenWebText</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.emnlp-main.41.mp4"/>
     </paper>
     <paper id="42">
       <title>Mitigating Language-Dependent Ethnic Bias in <fixed-case>BERT</fixed-case></title>
@@ -645,9 +645,9 @@
       <url hash="dca81d03">2021.emnlp-main.42</url>
       <bibkey>ahn-oh-2021-mitigating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.42</doi>
+      <video href="2021.emnlp-main.42.mp4"/>
       <pwccode url="https://github.com/jaimeenahn/ethnic_bias" additional="false">jaimeenahn/ethnic_bias</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.emnlp-main.42.mp4"/>
     </paper>
     <paper id="43">
       <title>Adversarial Scrubbing of Demographic Information for Text Classification</title>
@@ -663,8 +663,8 @@
       <attachment type="Software" hash="c3c357e6">2021.emnlp-main.43.Software.zip</attachment>
       <bibkey>basu-roy-chowdhury-etal-2021-adversarial</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.43</doi>
-      <pwccode url="https://github.com/brcsomnath/adversarial-scrubber" additional="false">brcsomnath/adversarial-scrubber</pwccode>
       <video href="2021.emnlp-main.43.mp4"/>
+      <pwccode url="https://github.com/brcsomnath/adversarial-scrubber" additional="false">brcsomnath/adversarial-scrubber</pwccode>
     </paper>
     <paper id="44">
       <title>Open-domain clarification question generation without question examples</title>
@@ -678,9 +678,9 @@
       <url hash="3eac608c">2021.emnlp-main.44</url>
       <bibkey>white-etal-2021-open</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.44</doi>
+      <video href="2021.emnlp-main.44.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/shapeworld">ShapeWorld</pwcdataset>
-      <video href="2021.emnlp-main.44.mp4"/>
     </paper>
     <paper id="45">
       <title>Improving Sequence-to-Sequence Pre-training via Sequence Span Rewriting</title>
@@ -694,8 +694,8 @@
       <url hash="9e568e70">2021.emnlp-main.45</url>
       <bibkey>zhou-etal-2021-improving-sequence</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.45</doi>
-      <pwccode url="https://github.com/michaelzhouwang/sequence_span_rewriting" additional="false">michaelzhouwang/sequence_span_rewriting</pwccode>
       <video href="2021.emnlp-main.45.mp4"/>
+      <pwccode url="https://github.com/michaelzhouwang/sequence_span_rewriting" additional="false">michaelzhouwang/sequence_span_rewriting</pwccode>
     </paper>
     <paper id="46">
       <title><fixed-case>C</fixed-case>oarse2<fixed-case>F</fixed-case>ine: Fine-grained Text Classification on Coarsely-grained Annotated Data</title>
@@ -719,9 +719,9 @@
       <url hash="973b90dd">2021.emnlp-main.47</url>
       <bibkey>edwards-etal-2021-text2mol</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.47</doi>
+      <video href="2021.emnlp-main.47.mp4"/>
       <pwccode url="https://github.com/cnedwards/text2mol" additional="false">cnedwards/text2mol</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/chebi-20">ChEBI-20</pwcdataset>
-      <video href="2021.emnlp-main.47.mp4"/>
     </paper>
     <paper id="48">
       <title>Classification of hierarchical text using geometric deep learning: the case of clinical trials corpus</title>
@@ -735,8 +735,8 @@
       <url hash="fd109006">2021.emnlp-main.48</url>
       <bibkey>ferdowsi-etal-2021-classification</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.48</doi>
-      <pwccode url="https://github.com/sssohrab/ct-classification-graphs" additional="false">sssohrab/ct-classification-graphs</pwccode>
       <video href="2021.emnlp-main.48.mp4"/>
+      <pwccode url="https://github.com/sssohrab/ct-classification-graphs" additional="false">sssohrab/ct-classification-graphs</pwccode>
     </paper>
     <paper id="49">
       <title>The Devil is in the Detail: Simple Tricks Improve Systematic Generalization of Transformers</title>
@@ -748,11 +748,11 @@
       <url hash="c1847c23">2021.emnlp-main.49</url>
       <bibkey>csordas-etal-2021-devil</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.49</doi>
+      <video href="2021.emnlp-main.49.mp4"/>
       <pwccode url="https://github.com/robertcsordas/transformer_generalization" additional="false">robertcsordas/transformer_generalization</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mathematics">Mathematics Dataset</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
-      <video href="2021.emnlp-main.49.mp4"/>
     </paper>
     <paper id="50">
       <title>Artificial Text Detection via Examining the Topology of Attention Maps</title>
@@ -770,10 +770,10 @@
       <url hash="60727d92">2021.emnlp-main.50</url>
       <bibkey>kushnareva-etal-2021-artificial</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.50</doi>
+      <video href="2021.emnlp-main.50.mp4"/>
       <pwccode url="https://github.com/danchern97/tda4atd" additional="false">danchern97/tda4atd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/realnews">RealNews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.emnlp-main.50.mp4"/>
     </paper>
     <paper id="51">
       <title>Active Learning by Acquiring Contrastive Examples</title>
@@ -786,13 +786,13 @@
       <url hash="593c8cf3">2021.emnlp-main.51</url>
       <bibkey>margatina-etal-2021-active</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.51</doi>
+      <video href="2021.emnlp-main.51.mp4"/>
       <pwccode url="https://github.com/mourga/contrastive-active-learning" additional="false">mourga/contrastive-active-learning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.51.mp4"/>
     </paper>
     <paper id="52">
       <title>Conditional <fixed-case>P</fixed-case>oisson Stochastic Beams</title>
@@ -826,8 +826,8 @@
       <url hash="110a97d3">2021.emnlp-main.53</url>
       <bibkey>batra-etal-2021-building</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.53</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/viggo">ViGGO</pwcdataset>
       <video href="2021.emnlp-main.53.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/viggo">ViGGO</pwcdataset>
     </paper>
     <paper id="54">
       <title>Moral Stories: Situated Reasoning about Norms, Intents, Actions, and their Consequences</title>
@@ -842,9 +842,9 @@
       <attachment type="Software" hash="75043371">2021.emnlp-main.54.Software.zip</attachment>
       <bibkey>emelin-etal-2021-moral</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.54</doi>
+      <video href="2021.emnlp-main.54.mp4"/>
       <pwccode url="https://github.com/demelin/moral_stories" additional="false">demelin/moral_stories</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/moral-stories">Moral Stories</pwcdataset>
-      <video href="2021.emnlp-main.54.mp4"/>
     </paper>
     <paper id="55">
       <title>Truth-Conditional Captions for Time Series Data</title>
@@ -855,8 +855,8 @@
       <url hash="0fe2cacc">2021.emnlp-main.55</url>
       <bibkey>jhamtani-berg-kirkpatrick-2021-truth</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.55</doi>
-      <pwccode url="https://github.com/harsh19/truce" additional="false">harsh19/truce</pwccode>
       <video href="2021.emnlp-main.55.mp4"/>
+      <pwccode url="https://github.com/harsh19/truce" additional="false">harsh19/truce</pwccode>
     </paper>
     <paper id="56">
       <title>Injecting Entity Types into Entity-Guided Text Generation</title>
@@ -869,8 +869,8 @@
       <url hash="1c5ea289">2021.emnlp-main.56</url>
       <bibkey>dong-etal-2021-injecting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.56</doi>
-      <pwccode url="https://github.com/wyu97/InjType" additional="true">wyu97/InjType</pwccode>
       <video href="2021.emnlp-main.56.mp4"/>
+      <pwccode url="https://github.com/wyu97/InjType" additional="true">wyu97/InjType</pwccode>
     </paper>
     <paper id="57">
       <title>Smelting Gold and Silver for Improved Multilingual <fixed-case>AMR</fixed-case>-to-<fixed-case>T</fixed-case>ext Generation</title>
@@ -883,8 +883,8 @@
       <url hash="3ae8143b">2021.emnlp-main.57</url>
       <bibkey>ribeiro-etal-2021-smelting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.57</doi>
-      <pwccode url="https://github.com/ukplab/m-amr2text" additional="false">ukplab/m-amr2text</pwccode>
       <video href="2021.emnlp-main.57.mp4"/>
+      <pwccode url="https://github.com/ukplab/m-amr2text" additional="false">ukplab/m-amr2text</pwccode>
     </paper>
     <paper id="58">
       <title>Learning Compact Metrics for <fixed-case>MT</fixed-case></title>
@@ -900,10 +900,10 @@
       <revision id="1" href="2021.emnlp-main.58v1" hash="def5905f"/>
       <revision id="2" href="2021.emnlp-main.58v2" hash="01c0de6f" date="2021-11-09">updated Figure 5 and Section 3</revision>
       <doi>10.18653/v1/2021.emnlp-main.58</doi>
+      <video href="2021.emnlp-main.58.mp4"/>
       <pwccode url="https://github.com/google-research/bleurt" additional="false">google-research/bleurt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mc4">mC4</pwcdataset>
-      <video href="2021.emnlp-main.58.mp4"/>
     </paper>
     <paper id="59">
       <title>The Impact of Positional Encodings on Multilingual Compression</title>
@@ -914,9 +914,9 @@
       <url hash="fb003bd3">2021.emnlp-main.59</url>
       <bibkey>ravishankar-sogaard-2021-impact</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.59</doi>
+      <video href="2021.emnlp-main.59.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.emnlp-main.59.mp4"/>
     </paper>
     <paper id="60">
       <title>Disentangling Representations of Text by Masking Transformers</title>
@@ -928,8 +928,8 @@
       <url hash="87689b47">2021.emnlp-main.60</url>
       <bibkey>zhang-etal-2021-disentangling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.60</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <video href="2021.emnlp-main.60.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
     </paper>
     <paper id="61">
       <title>Exploring the Role of <fixed-case>BERT</fixed-case> Token Representations to Explain Sentence Probing Results</title>
@@ -941,8 +941,8 @@
       <url hash="9b528f3a">2021.emnlp-main.61</url>
       <bibkey>mohebbi-etal-2021-exploring</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.61</doi>
-      <pwccode url="https://github.com/hmohebbi/explain-probing-results" additional="false">hmohebbi/explain-probing-results</pwccode>
       <video href="2021.emnlp-main.61.mp4"/>
+      <pwccode url="https://github.com/hmohebbi/explain-probing-results" additional="false">hmohebbi/explain-probing-results</pwccode>
     </paper>
     <paper id="62">
       <title>Do Long-Range Language Models Actually Use Long-Range Context?</title>
@@ -955,8 +955,8 @@
       <url hash="dde33f63">2021.emnlp-main.62</url>
       <bibkey>sun-etal-2021-long</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.62</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/pg-19">PG-19</pwcdataset>
       <video href="2021.emnlp-main.62.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/pg-19">PG-19</pwcdataset>
     </paper>
     <paper id="63">
       <title><fixed-case>T</fixed-case>he <fixed-case>W</fixed-case>orld of an <fixed-case>O</fixed-case>ctopus: <fixed-case>H</fixed-case>ow <fixed-case>R</fixed-case>eporting <fixed-case>B</fixed-case>ias <fixed-case>I</fixed-case>nfluences a <fixed-case>L</fixed-case>anguage <fixed-case>M</fixed-case>odel’s <fixed-case>P</fixed-case>erception of <fixed-case>C</fixed-case>olor</title>
@@ -982,9 +982,9 @@
       <url hash="edbdbccc">2021.emnlp-main.64</url>
       <bibkey>rajagopal-etal-2021-selfexplain</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.64</doi>
+      <video href="2021.emnlp-main.64.mp4"/>
       <pwccode url="https://github.com/dheerajrajagopal/SelfExplain" additional="true">dheerajrajagopal/SelfExplain</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.64.mp4"/>
     </paper>
     <paper id="65">
       <title>Memory and Knowledge Augmented Language Models for Inferring Salience in Long-Form Stories</title>
@@ -995,9 +995,9 @@
       <url hash="1518c71b">2021.emnlp-main.65</url>
       <bibkey>wilmot-keller-2021-memory</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.65</doi>
+      <video href="2021.emnlp-main.65.mp4"/>
       <pwccode url="https://github.com/dwlmt/story-fragments" additional="false">dwlmt/story-fragments</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/shmoop-corpus">Shmoop Corpus</pwcdataset>
-      <video href="2021.emnlp-main.65.mp4"/>
     </paper>
     <paper id="66">
       <title>Semantic Novelty Detection in Natural Language Descriptions</title>
@@ -1013,10 +1013,10 @@
       <url hash="62ac87aa">2021.emnlp-main.66</url>
       <bibkey>ma-etal-2021-semantic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.66</doi>
+      <video href="2021.emnlp-main.66.mp4"/>
       <pwccode url="https://github.com/nianzuma/semantic-novelty-detection-in-natural-language-descriptions" additional="false">nianzuma/semantic-novelty-detection-in-natural-language-descriptions</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
-      <video href="2021.emnlp-main.66.mp4"/>
     </paper>
     <paper id="67">
       <title>Jump-Starting Item Parameters for Adaptive Language Tests</title>
@@ -1064,8 +1064,8 @@
       <url hash="0fedef02">2021.emnlp-main.69</url>
       <bibkey>falis-etal-2021-cophe</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.69</doi>
-      <pwccode url="https://github.com/modr00cka/cophe" additional="false">modr00cka/cophe</pwccode>
       <video href="2021.emnlp-main.69.mp4"/>
+      <pwccode url="https://github.com/modr00cka/cophe" additional="false">modr00cka/cophe</pwccode>
     </paper>
     <paper id="70">
       <title>Learning Universal Authorship Representations</title>
@@ -1092,10 +1092,10 @@
       <url hash="4d22e181">2021.emnlp-main.71</url>
       <bibkey>yu-xu-2021-predicting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.71</doi>
+      <video href="2021.emnlp-main.71.mp4"/>
       <pwccode url="https://github.com/jadeleiyu/frame_extension" additional="false">jadeleiyu/frame_extension</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imagenet">ImageNet</pwcdataset>
-      <video href="2021.emnlp-main.71.mp4"/>
     </paper>
     <paper id="72">
       <title>Frequency Effects on Syntactic Rule Learning in Transformers</title>
@@ -1108,8 +1108,8 @@
       <url hash="eadf4986">2021.emnlp-main.72</url>
       <bibkey>wei-etal-2021-frequency</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.72</doi>
-      <pwccode url="https://github.com/google-research/language" additional="false">google-research/language</pwccode>
       <video href="2021.emnlp-main.72.mp4"/>
+      <pwccode url="https://github.com/google-research/language" additional="false">google-research/language</pwccode>
     </paper>
     <paper id="73">
       <title>A surprisal–duration trade-off across and within the world’s languages</title>
@@ -1140,10 +1140,11 @@
       <url hash="75220e7a">2021.emnlp-main.74</url>
       <bibkey>meister-etal-2021-revisiting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.74</doi>
+      <video href="2021.emnlp-main.74.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-stories">Natural Stories</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.emnlp-main.74.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="75">
       <title>Condenser: a Pre-training Architecture for Dense Retrieval</title>
@@ -1154,11 +1155,11 @@
       <url hash="6dae7ee9">2021.emnlp-main.75</url>
       <bibkey>gao-callan-2021-condenser</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.75</doi>
+      <video href="2021.emnlp-main.75.mp4"/>
       <pwccode url="https://github.com/luyug/Condenser" additional="false">luyug/Condenser</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.emnlp-main.75.mp4"/>
     </paper>
     <paper id="76">
       <title>Monitoring geometrical properties of word embeddings for detecting the emergence of new topics.</title>
@@ -1184,10 +1185,10 @@
       <url hash="8aa06882">2021.emnlp-main.77</url>
       <bibkey>lin-etal-2021-contextualized</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.77</doi>
+      <video href="2021.emnlp-main.77.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/canard">CANARD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/orconvqa">ORConvQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quac">QuAC</pwcdataset>
-      <video href="2021.emnlp-main.77.mp4"/>
     </paper>
     <paper id="78">
       <title>Ultra-High Dimensional Sparse Representations with Binarization for Efficient Text Retrieval</title>
@@ -1204,8 +1205,8 @@
       <attachment type="Software" hash="8230fe2a">2021.emnlp-main.78.Software.zip</attachment>
       <bibkey>jang-etal-2021-ultra</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.78</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <video href="2021.emnlp-main.78.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
     </paper>
     <paper id="79">
       <title><fixed-case>IR</fixed-case> like a <fixed-case>SIR</fixed-case>: <fixed-case>S</fixed-case>ense-enhanced <fixed-case>I</fixed-case>nformation <fixed-case>R</fixed-case>etrieval for <fixed-case>M</fixed-case>ultiple <fixed-case>L</fixed-case>anguages</title>
@@ -1220,8 +1221,8 @@
       <url hash="d8809a09">2021.emnlp-main.79</url>
       <bibkey>blloshmi-etal-2021-ir</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.79</doi>
-      <pwccode url="https://github.com/sapienzanlp/sir" additional="false">sapienzanlp/sir</pwccode>
       <video href="2021.emnlp-main.79.mp4"/>
+      <pwccode url="https://github.com/sapienzanlp/sir" additional="false">sapienzanlp/sir</pwccode>
     </paper>
     <paper id="80">
       <title>Neural Attention-Aware Hierarchical Topic Model</title>
@@ -1246,10 +1247,10 @@
       <url hash="e8384181">2021.emnlp-main.81</url>
       <bibkey>safavi-koutra-2021-relational</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.81</doi>
+      <video href="2021.emnlp-main.81.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
-      <video href="2021.emnlp-main.81.mp4"/>
     </paper>
     <paper id="82">
       <title>Certified Robustness to Programmable Transformations in <fixed-case>LSTM</fixed-case>s</title>
@@ -1277,11 +1278,11 @@
       <url hash="326c9605">2021.emnlp-main.83</url>
       <bibkey>dognin-etal-2021-regen</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.83</doi>
+      <video href="2021.emnlp-main.83.mp4"/>
       <pwccode url="https://github.com/IBM/regen" additional="false">IBM/regen</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kelm">KELM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tekgen">TekGen</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.emnlp-main.83.mp4"/>
     </paper>
     <paper id="84">
       <title>Contrastive Out-of-Distribution Detection for Pretrained Transformers</title>
@@ -1293,10 +1294,10 @@
       <url hash="5ebfdcb0">2021.emnlp-main.84</url>
       <bibkey>zhou-etal-2021-contrastive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.84</doi>
+      <video href="2021.emnlp-main.84.mp4"/>
       <pwccode url="https://github.com/wzhouad/Contra-OOD" additional="false">wzhouad/Contra-OOD</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2016">WMT 2016</pwcdataset>
-      <video href="2021.emnlp-main.84.mp4"/>
     </paper>
     <paper id="85">
       <title><fixed-case>M</fixed-case>ind<fixed-case>C</fixed-case>raft: Theory of Mind Modeling for Situated Dialogue in Collaborative Tasks</title>
@@ -1308,8 +1309,8 @@
       <url hash="f79e0dbd">2021.emnlp-main.85</url>
       <bibkey>bara-etal-2021-mindcraft</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.85</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/mindcraft">MindCraft</pwcdataset>
       <video href="2021.emnlp-main.85.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mindcraft">MindCraft</pwcdataset>
     </paper>
     <paper id="86">
       <title>Detecting Speaker Personas from Conversational Texts</title>
@@ -1324,10 +1325,10 @@
       <url hash="d5a7e646">2021.emnlp-main.86</url>
       <bibkey>gu-etal-2021-detecting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.86</doi>
+      <video href="2021.emnlp-main.86.mp4"/>
       <pwccode url="https://github.com/jasonforjoy/spd" additional="false">jasonforjoy/spd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/pmpc">PMPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.emnlp-main.86.mp4"/>
     </paper>
     <paper id="87">
       <title>Cross-lingual Intermediate Fine-tuning improves Dialogue State Tracking</title>
@@ -1339,9 +1340,9 @@
       <url hash="481ac43c">2021.emnlp-main.87</url>
       <bibkey>moghe-etal-2021-cross</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.87</doi>
+      <video href="2021.emnlp-main.87.mp4"/>
       <pwccode url="https://github.com/nikitacs16/xlift_dst" additional="false">nikitacs16/xlift_dst</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
-      <video href="2021.emnlp-main.87.mp4"/>
     </paper>
     <paper id="88">
       <title><fixed-case>ConvFiT:</fixed-case> <fixed-case>C</fixed-case>onversational Fine-Tuning of Pretrained Language Models</title>
@@ -1358,10 +1359,10 @@
       <url hash="7d2efc00">2021.emnlp-main.88</url>
       <bibkey>vulic-etal-2021-convfit</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.88</doi>
+      <video href="2021.emnlp-main.88.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/clinc150">CLINC150</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/reddit-corpus">Reddit Corpus</pwcdataset>
-      <video href="2021.emnlp-main.88.mp4"/>
     </paper>
     <paper id="89">
       <title>We’ve had this conversation before: A Novel Approach to Measuring Dialog Similarity</title>
@@ -1388,10 +1389,10 @@
       <url hash="9479ca2b">2021.emnlp-main.90</url>
       <bibkey>kahardipraja-etal-2021-towards</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.90</doi>
+      <video href="2021.emnlp-main.90.mp4"/>
       <pwccode url="https://github.com/pkhdipraja/towards-incremental-transformers" additional="false">pkhdipraja/towards-incremental-transformers</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.emnlp-main.90.mp4"/>
     </paper>
     <paper id="91">
       <title>Feedback Attribution for Counterfactual Bandit Learning in Multi-Domain Spoken Language Understanding</title>
@@ -1402,8 +1403,8 @@
       <url hash="db9b6c9c">2021.emnlp-main.91</url>
       <bibkey>falke-lehnen-2021-feedback</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.91</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
       <video href="2021.emnlp-main.91.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
     </paper>
     <paper id="92">
       <title>Label Verbalization and Entailment for Effective Zero and Few-Shot Relation Extraction</title>
@@ -1417,8 +1418,8 @@
       <url hash="26a706c7">2021.emnlp-main.92</url>
       <bibkey>sainz-etal-2021-label</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.92</doi>
-      <pwccode url="https://github.com/osainz59/Ask2Transformers" additional="false">osainz59/Ask2Transformers</pwccode>
       <video href="2021.emnlp-main.92.mp4"/>
+      <pwccode url="https://github.com/osainz59/Ask2Transformers" additional="false">osainz59/Ask2Transformers</pwccode>
     </paper>
     <paper id="93">
       <title>Extend, don’t rebuild: Phrasing conditional graph modification as autoregressive sequence labelling</title>
@@ -1447,11 +1448,11 @@
       <url hash="f1896a99">2021.emnlp-main.94</url>
       <bibkey>wang-etal-2021-zero</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.94</doi>
+      <video href="2021.emnlp-main.94.mp4"/>
       <pwccode url="https://github.com/cgraywang/deepex" additional="false">cgraywang/deepex</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tacred">TACRED</pwcdataset>
-      <video href="2021.emnlp-main.94.mp4"/>
     </paper>
     <paper id="95">
       <title>Learning Logic Rules for Document-Level Relation Extraction</title>
@@ -1468,10 +1469,10 @@
       <url hash="e34e0708">2021.emnlp-main.95</url>
       <bibkey>ru-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.95</doi>
+      <video href="2021.emnlp-main.95.mp4"/>
       <pwccode url="https://github.com/rudongyu/logire" additional="false">rudongyu/logire</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dwie">DWIE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/docred">DocRED</pwcdataset>
-      <video href="2021.emnlp-main.95.mp4"/>
     </paper>
     <paper id="96">
       <title>A Large-Scale Dataset for Empathetic Response Generation</title>
@@ -1484,11 +1485,11 @@
       <attachment type="Software" hash="aa44c511">2021.emnlp-main.96.Software.zip</attachment>
       <bibkey>welivita-etal-2021-large</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.96</doi>
+      <video href="2021.emnlp-main.96.mp4"/>
       <pwccode url="https://github.com/anuradha1992/edos" additional="false">anuradha1992/edos</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/emotionlines">EmotionLines</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
-      <video href="2021.emnlp-main.96.mp4"/>
     </paper>
     <paper id="97">
       <title>The Perils of Using <fixed-case>M</fixed-case>echanical <fixed-case>T</fixed-case>urk to Evaluate Open-Ended Text Generation</title>
@@ -1500,8 +1501,8 @@
       <url hash="2e2934c8">2021.emnlp-main.97</url>
       <bibkey>karpinska-etal-2021-perils</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.97</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
       <video href="2021.emnlp-main.97.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
     </paper>
     <paper id="98">
       <title>Documenting Large Webtext Corpora: A Case Study on the Colossal Clean Crawled Corpus</title>
@@ -1518,6 +1519,7 @@
       <url hash="5812bef4">2021.emnlp-main.98</url>
       <bibkey>dodge-etal-2021-documenting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.98</doi>
+      <video href="2021.emnlp-main.98.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
@@ -1525,7 +1527,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikibio">WikiBio</pwcdataset>
-      <video href="2021.emnlp-main.98.mp4"/>
     </paper>
     <paper id="99">
       <title><fixed-case>A</fixed-case>fro<fixed-case>MT</fixed-case>: Pretraining Strategies and Reproducible Benchmarks for Translation of 8 <fixed-case>A</fixed-case>frican Languages</title>
@@ -1538,9 +1539,9 @@
       <url hash="71fac500">2021.emnlp-main.99</url>
       <bibkey>reid-etal-2021-afromt</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.99</doi>
+      <video href="2021.emnlp-main.99.mp4"/>
       <pwccode url="https://github.com/machelreid/afromt" additional="false">machelreid/afromt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
-      <video href="2021.emnlp-main.99.mp4"/>
     </paper>
     <paper id="100">
       <title>Evaluating the Evaluation Metrics for Style Transfer: A Case Study in Multilingual Formality Transfer</title>
@@ -1553,10 +1554,10 @@
       <url hash="4b2d8333">2021.emnlp-main.100</url>
       <bibkey>briakou-etal-2021-evaluating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.100</doi>
+      <video href="2021.emnlp-main.100.mp4"/>
       <pwccode url="https://github.com/elbria/xformal-fost-meta" additional="false">elbria/xformal-fost-meta</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
-      <video href="2021.emnlp-main.100.mp4"/>
     </paper>
     <paper id="101">
       <title><fixed-case>MS</fixed-case>-Mentions: Consistently Annotating Entity Mentions in Materials Science Procedural Text</title>
@@ -1586,8 +1587,8 @@
       <attachment type="Software" hash="d114b293">2021.emnlp-main.102.Software.zip</attachment>
       <bibkey>pujari-goldwasser-2021-understanding</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.102</doi>
-      <pwccode url="https://github.com/pujari-rajkumar/compositional_learner" additional="false">pujari-rajkumar/compositional_learner</pwccode>
       <video href="2021.emnlp-main.102.mp4"/>
+      <pwccode url="https://github.com/pujari-rajkumar/compositional_learner" additional="false">pujari-rajkumar/compositional_learner</pwccode>
     </paper>
     <paper id="103">
       <title>Conundrums in Event Coreference Resolution: Making Sense of the State of the Art</title>
@@ -1612,8 +1613,8 @@
       <url hash="88cbc0fb">2021.emnlp-main.104</url>
       <bibkey>gravellier-etal-2021-weakly</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.104</doi>
-      <pwccode url="https://github.com/linto-project/linto-dialogue-act-segmentation" additional="false">linto-project/linto-dialogue-act-segmentation</pwccode>
       <video href="2021.emnlp-main.104.mp4"/>
+      <pwccode url="https://github.com/linto-project/linto-dialogue-act-segmentation" additional="false">linto-project/linto-dialogue-act-segmentation</pwccode>
     </paper>
     <paper id="105">
       <title>Narrative Embedding: <fixed-case>R</fixed-case>e-<fixed-case>C</fixed-case>ontextualization Through Attention</title>
@@ -1637,9 +1638,9 @@
       <url hash="289c46b6">2021.emnlp-main.106</url>
       <bibkey>held-etal-2021-focus</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.106</doi>
+      <video href="2021.emnlp-main.106.mp4"/>
       <pwccode url="https://github.com/helw150/event_entity_coref_ecb_plus" additional="false">helw150/event_entity_coref_ecb_plus</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ecb">ECB+</pwcdataset>
-      <video href="2021.emnlp-main.106.mp4"/>
     </paper>
     <paper id="107">
       <title>Salience-Aware Event Chain Modeling for Narrative Understanding</title>
@@ -1651,9 +1652,9 @@
       <url hash="fede07d4">2021.emnlp-main.107</url>
       <bibkey>zhang-etal-2021-salience</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.107</doi>
+      <video href="2021.emnlp-main.107.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/torque">Torque</pwcdataset>
-      <video href="2021.emnlp-main.107.mp4"/>
     </paper>
     <paper id="108">
       <title>Asking It All: Generating Contextualized Questions for any Semantic Role</title>
@@ -1668,12 +1669,12 @@
       <url hash="9d93bf70">2021.emnlp-main.108</url>
       <bibkey>pyatkin-etal-2021-asking</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.108</doi>
+      <video href="2021.emnlp-main.108.mp4"/>
       <pwccode url="https://github.com/valentinapy/roleqgeneration" additional="false">valentinapy/roleqgeneration</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/nombank">NomBank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qa-srl">QA-SRL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qa-srl-bank-2-0">QA-SRL Bank 2.0</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.108.mp4"/>
     </paper>
     <paper id="109">
       <title>Fast, Effective, and Self-Supervised: Transforming Masked Language Models into Universal Lexical and Sentence Encoders</title>
@@ -1686,6 +1687,7 @@
       <url hash="d1602175">2021.emnlp-main.109</url>
       <bibkey>liu-etal-2021-fast</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.109</doi>
+      <video href="2021.emnlp-main.109.mp4"/>
       <pwccode url="https://github.com/cambridgeltl/mirror-bert" additional="false">cambridgeltl/mirror-bert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cometa">COMETA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
@@ -1695,7 +1697,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sts-benchmark">STS Benchmark</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semantic-textual-similarity-2012-2016">Semantic Textual Similarity (2012 - 2016)</pwcdataset>
-      <video href="2021.emnlp-main.109.mp4"/>
     </paper>
     <paper id="110">
       <title><fixed-case>R</fixed-case>ule<fixed-case>BERT</fixed-case>: Teaching Soft Rules to Pre-Trained Language Models</title>
@@ -1708,10 +1709,9 @@
       <url hash="5569ca54">2021.emnlp-main.110</url>
       <bibkey>saeed-etal-2021-rulebert</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.110</doi>
+      <video href="2021.emnlp-main.110.mp4"/>
       <pwccode url="https://github.com/mhmdsaiid/rulebert" additional="false">mhmdsaiid/rulebert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
-      <pwcdataset url="https://paperswithcode.com/dataset/babi-1">bAbI</pwcdataset>
-      <video href="2021.emnlp-main.110.mp4"/>
     </paper>
     <paper id="111">
       <title>Stepmothers are mean and academics are pretentious: What do pretrained language models learn about you?</title>
@@ -1735,9 +1735,9 @@
       <url hash="24c093e6">2021.emnlp-main.112</url>
       <bibkey>barba-etal-2021-consec</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.112</doi>
+      <video href="2021.emnlp-main.112.mp4"/>
       <pwccode url="https://github.com/sapienzanlp/consec" additional="false">sapienzanlp/consec</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
-      <video href="2021.emnlp-main.112.mp4"/>
     </paper>
     <paper id="113">
       <title>Shortcutted Commonsense: Data Spuriousness in Deep Learning of Commonsense Reasoning</title>
@@ -1751,6 +1751,7 @@
       <attachment type="Software" hash="822c363a">2021.emnlp-main.113.Software.zip</attachment>
       <bibkey>branco-etal-2021-shortcutted</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.113</doi>
+      <video href="2021.emnlp-main.113.mp4"/>
       <pwccode url="https://github.com/nlx-group/shortcutted-commonsense-reasoning" additional="false">nlx-group/shortcutted-commonsense-reasoning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/arc">ARC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/commonsenseqa">CommonsenseQA</pwcdataset>
@@ -1758,7 +1759,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/piqa">PIQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.emnlp-main.113.mp4"/>
     </paper>
     <paper id="114">
       <title>When differential privacy meets <fixed-case>NLP</fixed-case>: The devil is in the detail</title>
@@ -1768,8 +1768,8 @@
       <url hash="cdc5bc96">2021.emnlp-main.114</url>
       <bibkey>habernal-2021-differential</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.114</doi>
-      <pwccode url="https://github.com/habernal/emnlp2021-differential-privacy-nlp" additional="false">habernal/emnlp2021-differential-privacy-nlp</pwccode>
       <video href="2021.emnlp-main.114.mp4"/>
+      <pwccode url="https://github.com/habernal/emnlp2021-differential-privacy-nlp" additional="false">habernal/emnlp2021-differential-privacy-nlp</pwccode>
     </paper>
     <paper id="115">
       <title>Achieving Model Robustness through Discrete Adversarial Training</title>
@@ -1780,11 +1780,11 @@
       <url hash="48e24533">2021.emnlp-main.115</url>
       <bibkey>ivgi-berant-2021-achieving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.115</doi>
+      <video href="2021.emnlp-main.115.mp4"/>
       <pwccode url="https://github.com/Mivg/robust_transformers" additional="false">Mivg/robust_transformers</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.115.mp4"/>
     </paper>
     <paper id="116">
       <title>Debiasing Methods in Natural Language Understanding Make Bias More Accessible</title>
@@ -1795,10 +1795,10 @@
       <url hash="52c518c3">2021.emnlp-main.116</url>
       <bibkey>mendelson-belinkov-2021-debiasing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.116</doi>
+      <video href="2021.emnlp-main.116.mp4"/>
       <pwccode url="https://github.com/technion-cs-nlp/bias-probing" additional="false">technion-cs-nlp/bias-probing</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.116.mp4"/>
     </paper>
     <paper id="117">
       <title>Evaluating the Robustness of Neural Language Models to Input Perturbations</title>
@@ -1809,9 +1809,9 @@
       <url hash="8789c234">2021.emnlp-main.117</url>
       <bibkey>moradi-samwald-2021-evaluating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.117</doi>
+      <video href="2021.emnlp-main.117.mp4"/>
       <pwccode url="https://github.com/mmoradi-iut/nlp-perturbation" additional="false">mmoradi-iut/nlp-perturbation</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.117.mp4"/>
     </paper>
     <paper id="118">
       <title>How much pretraining data do language models need to learn syntax?</title>
@@ -1823,8 +1823,8 @@
       <url hash="d7c4b99f">2021.emnlp-main.118</url>
       <bibkey>perez-mayos-etal-2021-much</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.118</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <video href="2021.emnlp-main.118.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
     </paper>
     <paper id="119">
       <title>Sorting through the noise: Testing robustness of information processing in pre-trained language models</title>
@@ -1850,10 +1850,10 @@
       <url hash="6c1403c5">2021.emnlp-main.120</url>
       <bibkey>jacovi-etal-2021-contrastive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.120</doi>
+      <video href="2021.emnlp-main.120.mp4"/>
       <pwccode url="https://github.com/allenai/contrastive-explanations" additional="false">allenai/contrastive-explanations</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.120.mp4"/>
     </paper>
     <paper id="121">
       <title>On the Transferability of Adversarial Attacks against Neural Text Classifier</title>
@@ -1867,9 +1867,9 @@
       <url hash="c0faf0f7">2021.emnlp-main.121</url>
       <bibkey>yuan-etal-2021-transferability</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.121</doi>
+      <video href="2021.emnlp-main.121.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.121.mp4"/>
     </paper>
     <paper id="122">
       <title>Conditional probing: measuring usable information beyond a baseline</title>
@@ -1882,9 +1882,9 @@
       <url hash="f129b85e">2021.emnlp-main.122</url>
       <bibkey>hewitt-etal-2021-conditional</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.122</doi>
+      <video href="2021.emnlp-main.122.mp4"/>
       <pwccode url="https://github.com/john-hewitt/conditional-probing" additional="false">john-hewitt/conditional-probing</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.122.mp4"/>
     </paper>
     <paper id="123">
       <title><fixed-case>GFST</fixed-case>: <fixed-case>G</fixed-case>ender-Filtered Self-Training for More Accurate Gender in Translation</title>
@@ -1897,8 +1897,8 @@
       <url hash="22313a5b">2021.emnlp-main.123</url>
       <bibkey>choubey-etal-2021-gfst</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.123</doi>
-      <pwccode url="https://github.com/amazon-research/gfst-nmt" additional="false">amazon-research/gfst-nmt</pwccode>
       <video href="2021.emnlp-main.123.mp4"/>
+      <pwccode url="https://github.com/amazon-research/gfst-nmt" additional="false">amazon-research/gfst-nmt</pwccode>
     </paper>
     <paper id="124">
       <title>“Wikily” Supervised Neural Translation Tailored to Cross-Lingual Tasks</title>
@@ -1910,8 +1910,8 @@
       <url hash="6485cc00">2021.emnlp-main.124</url>
       <bibkey>rasooli-etal-2021-wikily</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.124</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <video href="2021.emnlp-main.124.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
     </paper>
     <paper id="125">
       <title>m<fixed-case>T</fixed-case>6: Multilingual Pretrained Text-to-Text Transformer with Translation Pairs</title>
@@ -1929,13 +1929,13 @@
       <url hash="080a730a">2021.emnlp-main.125</url>
       <bibkey>chi-etal-2021-mt6</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.125</doi>
+      <video href="2021.emnlp-main.125.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikilingua">WikiLingua</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.emnlp-main.125.mp4"/>
     </paper>
     <paper id="126">
       <title>Improving Zero-Shot Cross-Lingual Transfer Learning via Robust Training</title>
@@ -1948,10 +1948,10 @@
       <url hash="0eff3601">2021.emnlp-main.126</url>
       <bibkey>huang-etal-2021-improving-zero</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.126</doi>
+      <video href="2021.emnlp-main.126.mp4"/>
       <pwccode url="https://github.com/uclanlp/robust-xlt" additional="false">uclanlp/robust-xlt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.emnlp-main.126.mp4"/>
     </paper>
     <paper id="127">
       <title>Speechformer: Reducing Information Loss in Direct Speech Translation</title>
@@ -1964,9 +1964,9 @@
       <url hash="a52d3ec5">2021.emnlp-main.127</url>
       <bibkey>papi-etal-2021-speechformer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.127</doi>
+      <video href="2021.emnlp-main.127.mp4"/>
       <pwccode url="https://github.com/sarapapi/fbk-fairseq" additional="false">sarapapi/fbk-fairseq</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
-      <video href="2021.emnlp-main.127.mp4"/>
     </paper>
     <paper id="128">
       <title>Is “moby dick” a Whale or a Bird? Named Entities and Terminology in Speech Translation</title>
@@ -1980,8 +1980,8 @@
       <url hash="3fe143fd">2021.emnlp-main.128</url>
       <bibkey>gaido-etal-2021-moby</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.128</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/europarl-st">Europarl-ST</pwcdataset>
       <video href="2021.emnlp-main.128.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/europarl-st">Europarl-ST</pwcdataset>
     </paper>
     <paper id="129">
       <title><fixed-case>H</fixed-case>inted<fixed-case>BT</fixed-case>: <fixed-case>A</fixed-case>ugmenting <fixed-case>B</fixed-case>ack-<fixed-case>T</fixed-case>ranslation with Quality and Transliteration Hints</title>
@@ -2006,8 +2006,8 @@
       <url hash="283b04f8">2021.emnlp-main.130</url>
       <bibkey>alinejad-etal-2021-translation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.130</doi>
-      <pwccode url="https://github.com/sfu-natlang/supervised-simultaneous-mt" additional="false">sfu-natlang/supervised-simultaneous-mt</pwccode>
       <video href="2021.emnlp-main.130.mp4"/>
+      <pwccode url="https://github.com/sfu-natlang/supervised-simultaneous-mt" additional="false">sfu-natlang/supervised-simultaneous-mt</pwccode>
     </paper>
     <paper id="131">
       <title>Nearest Neighbour Few-Shot Learning for Cross-lingual Classification</title>
@@ -2019,8 +2019,8 @@
       <url hash="a06e5a09">2021.emnlp-main.131</url>
       <bibkey>bari-etal-2021-nearest</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.131</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <video href="2021.emnlp-main.131.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
     </paper>
     <paper id="132">
       <title>Cross-Attention is All You Need: <fixed-case>A</fixed-case>dapting Pretrained <fixed-case>T</fixed-case>ransformers for Machine Translation</title>
@@ -2032,8 +2032,8 @@
       <url hash="d256e97e">2021.emnlp-main.132</url>
       <bibkey>gheini-etal-2021-cross</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.132</doi>
-      <pwccode url="https://github.com/mgheini/xattn-transfer-for-mt" additional="false">mgheini/xattn-transfer-for-mt</pwccode>
       <video href="2021.emnlp-main.132.mp4"/>
+      <pwccode url="https://github.com/mgheini/xattn-transfer-for-mt" additional="false">mgheini/xattn-transfer-for-mt</pwccode>
     </paper>
     <paper id="133">
       <title>Effects of Parameter Norm Growth During Transformer Training: Inductive Bias from Gradient Descent</title>
@@ -2047,10 +2047,10 @@
       <url hash="bdb4b222">2021.emnlp-main.133</url>
       <bibkey>merrill-etal-2021-effects</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.133</doi>
+      <video href="2021.emnlp-main.133.mp4"/>
       <pwccode url="https://github.com/viking-sudo-rm/norm-growth" additional="false">viking-sudo-rm/norm-growth</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
-      <video href="2021.emnlp-main.133.mp4"/>
     </paper>
     <paper id="134">
       <title>Foreseeing the Benefits of Incidental Supervision</title>
@@ -2063,10 +2063,10 @@
       <url hash="b0c3ac2b">2021.emnlp-main.134</url>
       <bibkey>he-etal-2021-foreseeing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.134</doi>
+      <video href="2021.emnlp-main.134.mp4"/>
       <pwccode url="https://github.com/CogComp/PABI" additional="true">CogComp/PABI</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/qa-srl-bank-2-0">QA-SRL Bank 2.0</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.134.mp4"/>
     </paper>
     <paper id="135">
       <title>Competency Problems: On Finding and Removing Artifacts in Language Data</title>
@@ -2083,12 +2083,12 @@
       <attachment type="Software" hash="089b5fb0">2021.emnlp-main.135.Software.tgz</attachment>
       <bibkey>gardner-etal-2021-competency</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.135</doi>
+      <video href="2021.emnlp-main.135.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/english-web-treebank">English Web Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.emnlp-main.135.mp4"/>
     </paper>
     <paper id="136">
       <title>Knowledge-Aware Meta-learning for Low-Resource Text Classification</title>
@@ -2101,8 +2101,8 @@
       <url hash="5e07393f">2021.emnlp-main.136</url>
       <bibkey>yao-etal-2021-knowledge</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.136</doi>
-      <pwccode url="https://github.com/huaxiuyao/KGML" additional="false">huaxiuyao/KGML</pwccode>
       <video href="2021.emnlp-main.136.mp4"/>
+      <pwccode url="https://github.com/huaxiuyao/KGML" additional="false">huaxiuyao/KGML</pwccode>
     </paper>
     <paper id="137">
       <title>Sentence Bottleneck Autoencoders from Transformer Language Models</title>
@@ -2114,9 +2114,9 @@
       <url hash="8bde0262">2021.emnlp-main.137</url>
       <bibkey>montero-etal-2021-sentence</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.137</doi>
+      <video href="2021.emnlp-main.137.mp4"/>
       <pwccode url="https://github.com/ivanmontero/autobot" additional="false">ivanmontero/autobot</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
-      <video href="2021.emnlp-main.137.mp4"/>
     </paper>
     <paper id="138">
       <title>Efficient Contrastive Learning via Novel Data Augmentation and Curriculum Learning</title>
@@ -2129,8 +2129,8 @@
       <attachment type="Software" hash="cb2d62a0">2021.emnlp-main.138.Software.zip</attachment>
       <bibkey>ye-etal-2021-efficient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.138</doi>
-      <pwccode url="https://github.com/vano1205/efficientcl" additional="false">vano1205/efficientcl</pwccode>
       <video href="2021.emnlp-main.138.mp4"/>
+      <pwccode url="https://github.com/vano1205/efficientcl" additional="false">vano1205/efficientcl</pwccode>
     </paper>
     <paper id="139">
       <title><fixed-case>CR</fixed-case>-Walker: Tree-Structured Graph Reasoning and Dialog Acts for Conversational Recommendation</title>
@@ -2142,10 +2142,10 @@
       <url hash="f5bcd32e">2021.emnlp-main.139</url>
       <bibkey>ma-etal-2021-cr</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.139</doi>
+      <video href="2021.emnlp-main.139.mp4"/>
       <pwccode url="https://github.com/truthless11/CR-Walker" additional="false">truthless11/CR-Walker</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/redial">ReDial</pwcdataset>
-      <video href="2021.emnlp-main.139.mp4"/>
     </paper>
     <paper id="140">
       <title><fixed-case>DIALKI</fixed-case>: Knowledge Identification in Conversational Systems through Dialogue-Document Contextualization</title>
@@ -2158,12 +2158,12 @@
       <url hash="e3d5606e">2021.emnlp-main.140</url>
       <bibkey>wu-etal-2021-dialki</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.140</doi>
+      <video href="2021.emnlp-main.140.mp4"/>
       <pwccode url="https://github.com/ellenmellon/dialki" additional="false">ellenmellon/dialki</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/doc2dial-1">Doc2Dial</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/holl-e">Holl-E</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/doc2dial">doc2dial</pwcdataset>
-      <video href="2021.emnlp-main.140.mp4"/>
     </paper>
     <paper id="141">
       <title>Iconary: A Pictionary-Based Game for Testing Multimodal Communication with Drawings and Text</title>
@@ -2188,9 +2188,9 @@
       <url hash="97743190">2021.emnlp-main.141</url>
       <bibkey>clark-etal-2021-iconary</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.141</doi>
+      <video href="2021.emnlp-main.141.mp4"/>
       <pwccode url="https://github.com/allenai/iconary" additional="false">allenai/iconary</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/iconary">Iconary</pwcdataset>
-      <video href="2021.emnlp-main.141.mp4"/>
     </paper>
     <paper id="142">
       <title>Self-training Improves Pre-training for Few-shot Learning in Task-oriented Dialog Systems</title>
@@ -2205,8 +2205,8 @@
       <url hash="13353330">2021.emnlp-main.142</url>
       <bibkey>mi-etal-2021-self</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.142</doi>
-      <pwccode url="https://github.com/mifei/st-tod" additional="false">mifei/st-tod</pwccode>
       <video href="2021.emnlp-main.142.mp4"/>
+      <pwccode url="https://github.com/mifei/st-tod" additional="false">mifei/st-tod</pwccode>
     </paper>
     <paper id="143">
       <title>Contextual Rephrase Detection for Reducing Friction in Dialogue Systems</title>
@@ -2240,8 +2240,8 @@
       <url hash="35bcd944">2021.emnlp-main.144</url>
       <bibkey>zhang-etal-2021-shot</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.144</doi>
-      <pwccode url="https://github.com/jianguoz/Few-Shot-Intent-Detection" additional="false">jianguoz/Few-Shot-Intent-Detection</pwccode>
       <video href="2021.emnlp-main.144.mp4"/>
+      <pwccode url="https://github.com/jianguoz/Few-Shot-Intent-Detection" additional="false">jianguoz/Few-Shot-Intent-Detection</pwccode>
     </paper>
     <paper id="145">
       <title>“It doesn’t look good for a date”: Transforming Critiques into Preferences for Conversational Recommendation Systems</title>
@@ -2268,8 +2268,8 @@
       <url hash="b8c7b71d">2021.emnlp-main.146</url>
       <bibkey>ding-luo-2021-attentionrank</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.146</doi>
-      <pwccode url="https://github.com/hd10-iupui/attentionrank" additional="false">hd10-iupui/attentionrank</pwccode>
       <video href="2021.emnlp-main.146.mp4"/>
+      <pwccode url="https://github.com/hd10-iupui/attentionrank" additional="false">hd10-iupui/attentionrank</pwccode>
     </paper>
     <paper id="147">
       <title>Unsupervised Relation Extraction: A Variational Autoencoder Approach</title>
@@ -2293,11 +2293,11 @@
       <url hash="9c7ab772">2021.emnlp-main.148</url>
       <bibkey>glass-etal-2021-robust</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.148</doi>
+      <video href="2021.emnlp-main.148.mp4"/>
       <pwccode url="https://github.com/ibm/kgi-slot-filling" additional="true">ibm/kgi-slot-filling</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
-      <video href="2021.emnlp-main.148.mp4"/>
     </paper>
     <paper id="149">
       <title>Everything Is All It Takes: A Multipronged Strategy for Zero-Shot Cross-Lingual Information Extraction</title>
@@ -2319,8 +2319,8 @@
       <url hash="3d756100">2021.emnlp-main.149</url>
       <bibkey>yarmohammadi-etal-2021-everything</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.149</doi>
-      <pwccode url="https://github.com/shijie-wu/crosslingual-nlp" additional="true">shijie-wu/crosslingual-nlp</pwccode>
       <video href="2021.emnlp-main.149.mp4"/>
+      <pwccode url="https://github.com/shijie-wu/crosslingual-nlp" additional="true">shijie-wu/crosslingual-nlp</pwccode>
     </paper>
     <paper id="150">
       <title>Harms of Gender Exclusivity and Challenges in Non-Binary Representation in Language Technologies</title>
@@ -2348,11 +2348,11 @@
       <attachment type="Software" hash="ceabaa30">2021.emnlp-main.151.Software.zip</attachment>
       <bibkey>wang-etal-2021-gender</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.151</doi>
+      <video href="2021.emnlp-main.151.mp4"/>
       <pwccode url="https://github.com/kuanghuei/SCAN" additional="false">kuanghuei/SCAN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coco-captions">COCO Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
-      <video href="2021.emnlp-main.151.mp4"/>
     </paper>
     <paper id="152">
       <title>Style Pooling: Automatic Text Style Obfuscation for Improved Classification Fairness</title>
@@ -2363,8 +2363,8 @@
       <url hash="e3860074">2021.emnlp-main.152</url>
       <bibkey>mireshghallah-berg-kirkpatrick-2021-style</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.152</doi>
-      <pwccode url="https://github.com/mireshghallah/style-pooling" additional="false">mireshghallah/style-pooling</pwccode>
       <video href="2021.emnlp-main.152.mp4"/>
+      <pwccode url="https://github.com/mireshghallah/style-pooling" additional="false">mireshghallah/style-pooling</pwccode>
     </paper>
     <paper id="153">
       <title>Modeling Disclosive Transparency in <fixed-case>NLP</fixed-case> Application Descriptions</title>
@@ -2378,8 +2378,8 @@
       <url hash="d4cf9b97">2021.emnlp-main.153</url>
       <bibkey>saxon-etal-2021-modeling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.153</doi>
-      <pwccode url="https://github.com/michaelsaxon/disclosive-transparency" additional="false">michaelsaxon/disclosive-transparency</pwccode>
       <video href="2021.emnlp-main.153.mp4"/>
+      <pwccode url="https://github.com/michaelsaxon/disclosive-transparency" additional="false">michaelsaxon/disclosive-transparency</pwccode>
     </paper>
     <paper id="154">
       <title>Reconstruction Attack on Instance Encoding for Language Understanding</title>
@@ -2390,9 +2390,9 @@
       <url hash="9e4baf94">2021.emnlp-main.154</url>
       <bibkey>xie-hong-2021-reconstruction</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.154</doi>
+      <video href="2021.emnlp-main.154.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.154.mp4"/>
     </paper>
     <paper id="155">
       <title>Fairness-aware Class Imbalanced Learning</title>
@@ -2419,8 +2419,8 @@
       <url hash="c9d13de1">2021.emnlp-main.156</url>
       <bibkey>feng-etal-2021-cryptogru</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.156</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <video href="2021.emnlp-main.156.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
     </paper>
     <paper id="157">
       <title>Local Word Discovery for Interactive Transcription</title>
@@ -2459,8 +2459,8 @@
       <url hash="e13448e7">2021.emnlp-main.159</url>
       <bibkey>goldman-tsarfaty-2021-minimal</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.159</doi>
-      <pwccode url="https://github.com/onlplab/morphodetection" additional="false">onlplab/morphodetection</pwccode>
       <video href="2021.emnlp-main.159.mp4"/>
+      <pwccode url="https://github.com/onlplab/morphodetection" additional="false">onlplab/morphodetection</pwccode>
     </paper>
     <paper id="160">
       <title>Fast <fixed-case>W</fixed-case>ord<fixed-case>P</fixed-case>iece Tokenization</title>
@@ -2485,8 +2485,8 @@
       <url hash="2fbc818c">2021.emnlp-main.161</url>
       <bibkey>cao-rimell-2021-evaluate</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.161</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/mc4">mC4</pwcdataset>
       <video href="2021.emnlp-main.161.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mc4">mC4</pwcdataset>
     </paper>
     <paper id="162">
       <title>Broaden the Vision: Geo-Diverse Visual Commonsense Reasoning</title>
@@ -2500,10 +2500,10 @@
       <url hash="af1525c9">2021.emnlp-main.162</url>
       <bibkey>yin-etal-2021-broaden</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.162</doi>
+      <video href="2021.emnlp-main.162.mp4"/>
       <pwccode url="https://github.com/wadeyin9712/gd-vcr" additional="false">wadeyin9712/gd-vcr</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gd-vcr">GD-VCR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vcr">VCR</pwcdataset>
-      <video href="2021.emnlp-main.162.mp4"/>
     </paper>
     <paper id="163">
       <title>Reference-Centric Models for Grounded Collaborative Dialogue</title>
@@ -2515,8 +2515,8 @@
       <url hash="6d204d3c">2021.emnlp-main.163</url>
       <bibkey>fried-etal-2021-reference</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.163</doi>
-      <pwccode url="https://github.com/dpfried/onecommon" additional="false">dpfried/onecommon</pwccode>
       <video href="2021.emnlp-main.163.mp4"/>
+      <pwccode url="https://github.com/dpfried/onecommon" additional="false">dpfried/onecommon</pwccode>
     </paper>
     <paper id="164">
       <title><fixed-case>C</fixed-case>ross<fixed-case>VQA</fixed-case>: Scalably Generating Benchmarks for Systematically Testing <fixed-case>VQA</fixed-case> Generalization</title>
@@ -2531,6 +2531,7 @@
       <url hash="cd78b493">2021.emnlp-main.164</url>
       <bibkey>akula-etal-2021-crossvqa</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.164</doi>
+      <video href="2021.emnlp-main.164.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gqa">GQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vcr">VCR</pwcdataset>
@@ -2538,7 +2539,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering-v2-0">Visual Question Answering v2.0</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vizwiz">VizWiz</pwcdataset>
-      <video href="2021.emnlp-main.164.mp4"/>
     </paper>
     <paper id="165">
       <title>Visual Goal-Step Inference using wiki<fixed-case>H</fixed-case>ow</title>
@@ -2553,11 +2553,11 @@
       <url hash="307192e8">2021.emnlp-main.165</url>
       <bibkey>yang-etal-2021-visual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.165</doi>
+      <video href="2021.emnlp-main.165.mp4"/>
       <pwccode url="https://github.com/yueyang1996/wikihow-vgsi" additional="false">yueyang1996/wikihow-vgsi</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihow-image">wikiHow-image</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coin">COIN</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/howto100m">HowTo100M</pwcdataset>
-      <video href="2021.emnlp-main.165.mp4"/>
     </paper>
     <paper id="166">
       <title>Systematic Generalization on g<fixed-case>SCAN</fixed-case>: <fixed-case>W</fixed-case>hat is Nearly Solved and What is Next?</title>
@@ -2571,9 +2571,9 @@
       <url hash="7fb6c527">2021.emnlp-main.166</url>
       <bibkey>qiu-etal-2021-systematic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.166</doi>
+      <video href="2021.emnlp-main.166.mp4"/>
       <pwccode url="https://github.com/LauraRuis/groundedSCAN" additional="false">LauraRuis/groundedSCAN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
-      <video href="2021.emnlp-main.166.mp4"/>
     </paper>
     <paper id="167">
       <title>Effect of Visual Extensions on Natural Language Understanding in Vision-and-Language Models</title>
@@ -2584,13 +2584,13 @@
       <url hash="92545804">2021.emnlp-main.167</url>
       <bibkey>iki-aizawa-2021-effect</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.167</doi>
+      <video href="2021.emnlp-main.167.mp4"/>
       <pwccode url="https://github.com/alab-nii/eval_vl_glue" additional="false">alab-nii/eval_vl_glue</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.167.mp4"/>
     </paper>
     <paper id="168">
       <title>Neural Path Hunter: Reducing Hallucination in Dialogue Systems via Path Grounding</title>
@@ -2603,9 +2603,9 @@
       <url hash="68faeceb">2021.emnlp-main.168</url>
       <bibkey>dziri-etal-2021-neural</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.168</doi>
+      <video href="2021.emnlp-main.168.mp4"/>
       <pwccode url="https://github.com/nouhadziri/neural-path-hunter" additional="false">nouhadziri/neural-path-hunter</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/opendialkg">OpenDialKG</pwcdataset>
-      <video href="2021.emnlp-main.168.mp4"/>
     </paper>
     <paper id="169">
       <title>Thinking Clearly, Talking Fast: Concept-Guided Non-Autoregressive Generation for Open-Domain Dialogue Systems</title>
@@ -2618,9 +2618,9 @@
       <url hash="24cc0603">2021.emnlp-main.169</url>
       <bibkey>zou-etal-2021-thinking</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.169</doi>
+      <video href="2021.emnlp-main.169.mp4"/>
       <pwccode url="https://github.com/rowitzou/cg-nar" additional="false">rowitzou/cg-nar</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.emnlp-main.169.mp4"/>
     </paper>
     <paper id="170">
       <title>Perspective-taking and Pragmatics for Generating Empathetic Responses Focused on Emotion Causes</title>
@@ -2632,11 +2632,11 @@
       <url hash="ea891378">2021.emnlp-main.170</url>
       <bibkey>kim-etal-2021-perspective</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.170</doi>
+      <video href="2021.emnlp-main.170.mp4"/>
       <pwccode url="https://github.com/skywalker023/focused-empathy" additional="false">skywalker023/focused-empathy</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/emocause">EmoCause</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/empatheticdialogues">EmpatheticDialogues</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/reccon">RECCON</pwcdataset>
-      <video href="2021.emnlp-main.170.mp4"/>
     </paper>
     <paper id="171">
       <title>Generation and Extraction Combined Dialogue State Tracking with Hierarchical Ontology Integration</title>
@@ -2649,9 +2649,9 @@
       <url hash="013d01b9">2021.emnlp-main.171</url>
       <bibkey>li-etal-2021-generation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.171</doi>
+      <video href="2021.emnlp-main.171.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/crosswoz">CrossWOZ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
-      <video href="2021.emnlp-main.171.mp4"/>
     </paper>
     <paper id="172">
       <title><fixed-case>C</fixed-case>o<fixed-case>LV</fixed-case>: A Collaborative Latent Variable Model for Knowledge-Grounded Dialogue Generation</title>
@@ -2664,9 +2664,9 @@
       <url hash="a5ef91da">2021.emnlp-main.172</url>
       <bibkey>zhan-etal-2021-colv</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.172</doi>
+      <video href="2021.emnlp-main.172.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/holl-e">Holl-E</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.emnlp-main.172.mp4"/>
     </paper>
     <paper id="173">
       <title><fixed-case>A</fixed-case> <fixed-case>T</fixed-case>hree-<fixed-case>S</fixed-case>tage <fixed-case>L</fixed-case>earning <fixed-case>F</fixed-case>ramework for <fixed-case>L</fixed-case>ow-<fixed-case>R</fixed-case>esource <fixed-case>K</fixed-case>nowledge-<fixed-case>G</fixed-case>rounded <fixed-case>D</fixed-case>ialogue <fixed-case>G</fixed-case>eneration</title>
@@ -2681,8 +2681,8 @@
       <url hash="6a74ba56">2021.emnlp-main.173</url>
       <bibkey>liu-etal-2021-three</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.173</doi>
-      <pwccode url="https://github.com/neukg/kat-tslf" additional="false">neukg/kat-tslf</pwccode>
       <video href="2021.emnlp-main.173.mp4"/>
+      <pwccode url="https://github.com/neukg/kat-tslf" additional="false">neukg/kat-tslf</pwccode>
     </paper>
     <paper id="174">
       <title>Intention Reasoning Network for Multi-Domain End-to-end Task-Oriented Dialogue</title>
@@ -2711,9 +2711,9 @@
       <url hash="85bea459">2021.emnlp-main.175</url>
       <bibkey>wu-etal-2021-better</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.175</doi>
+      <video href="2021.emnlp-main.175.mp4"/>
       <pwccode url="https://github.com/pku-sixing/emnlp2021-mske_dialog" additional="false">pku-sixing/emnlp2021-mske_dialog</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
-      <video href="2021.emnlp-main.175.mp4"/>
     </paper>
     <paper id="176">
       <title>Domain-Lifelong Learning for Dialogue State Tracking via Knowledge Preservation Networks</title>
@@ -2731,9 +2731,9 @@
       <url hash="68d73367">2021.emnlp-main.176</url>
       <bibkey>liu-etal-2021-domain</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.176</doi>
+      <video href="2021.emnlp-main.176.mp4"/>
       <pwccode url="https://github.com/liuqingbin/knowledge-preservation-networks" additional="false">liuqingbin/knowledge-preservation-networks</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sgd">SGD</pwcdataset>
-      <video href="2021.emnlp-main.176.mp4"/>
     </paper>
     <paper id="177">
       <title><fixed-case>CSAGN</fixed-case>: Conversational Structure Aware Graph Network for Conversational Semantic Role Labeling</title>
@@ -2745,8 +2745,8 @@
       <url hash="62b18cd4">2021.emnlp-main.177</url>
       <bibkey>wu-etal-2021-csagn</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.177</doi>
-      <pwccode url="https://github.com/hahahawu/CSAGN" additional="false">hahahawu/CSAGN</pwccode>
       <video href="2021.emnlp-main.177.mp4"/>
+      <pwccode url="https://github.com/hahahawu/CSAGN" additional="false">hahahawu/CSAGN</pwccode>
     </paper>
     <paper id="178">
       <title>Different Strokes for Different Folks: Investigating Appropriate Further Pre-training Approaches for Diverse Dialogue Tasks</title>
@@ -2771,9 +2771,9 @@
       <url hash="d5894ffa">2021.emnlp-main.179</url>
       <bibkey>cui-etal-2021-knowledge</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.179</doi>
+      <video href="2021.emnlp-main.179.mp4"/>
       <pwccode url="https://github.com/nealcly/ke-blender" additional="false">nealcly/ke-blender</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.emnlp-main.179.mp4"/>
     </paper>
     <paper id="180">
       <title>An Evaluation Dataset and Strategy for Building Robust Multi-turn Response Selection Model</title>
@@ -2785,8 +2785,8 @@
       <url hash="0da7574f">2021.emnlp-main.180</url>
       <bibkey>han-etal-2021-evaluation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.180</doi>
-      <pwccode url="https://github.com/kakaoenterprise/koradvmrstestdata" additional="false">kakaoenterprise/koradvmrstestdata</pwccode>
       <video href="2021.emnlp-main.180.mp4"/>
+      <pwccode url="https://github.com/kakaoenterprise/koradvmrstestdata" additional="false">kakaoenterprise/koradvmrstestdata</pwccode>
     </paper>
     <paper id="181">
       <title>Unsupervised Conversation Disentanglement through Co-Training</title>
@@ -2798,8 +2798,8 @@
       <url hash="4afe3bd1">2021.emnlp-main.181</url>
       <bibkey>liu-etal-2021-unsupervised</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.181</doi>
-      <pwccode url="https://github.com/layneins/unsupervised_dialo_disentanglement" additional="false">layneins/unsupervised_dialo_disentanglement</pwccode>
       <video href="2021.emnlp-main.181.mp4"/>
+      <pwccode url="https://github.com/layneins/unsupervised_dialo_disentanglement" additional="false">layneins/unsupervised_dialo_disentanglement</pwccode>
     </paper>
     <paper id="182">
       <title>Don’t be Contradicted with Anything! <fixed-case>CI</fixed-case>-<fixed-case>T</fixed-case>o<fixed-case>D</fixed-case>: Towards Benchmarking Consistency for Task-oriented Dialogue System</title>
@@ -2814,8 +2814,8 @@
       <url hash="ab1c6973">2021.emnlp-main.182</url>
       <bibkey>qin-etal-2021-dont</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.182</doi>
-      <pwccode url="https://github.com/yizhen20133868/ci-tod" additional="false">yizhen20133868/ci-tod</pwccode>
       <video href="2021.emnlp-main.182.mp4"/>
+      <pwccode url="https://github.com/yizhen20133868/ci-tod" additional="false">yizhen20133868/ci-tod</pwccode>
     </paper>
     <paper id="183">
       <title>Transferable Persona-Grounded Dialogues via Grounded Minimal Edits</title>
@@ -2829,6 +2829,7 @@
       <bibkey>wu-etal-2021-transferable</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.183</doi>
       <video href="2021.emnlp-main.183.mp4"/>
+      <pwccode url="https://github.com/thu-coai/grounded-minimal-edit" additional="false">thu-coai/grounded-minimal-edit</pwccode>
     </paper>
     <paper id="184">
       <title><fixed-case>EARL</fixed-case>: Informative Knowledge-Grounded Conversation Generation with Entity-Agnostic Representation Learning</title>
@@ -2842,9 +2843,9 @@
       <url hash="257d476e">2021.emnlp-main.184</url>
       <bibkey>zhou-etal-2021-earl</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.184</doi>
+      <video href="2021.emnlp-main.184.mp4"/>
       <pwccode url="https://github.com/thu-coai/earl" additional="false">thu-coai/earl</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/opendialkg">OpenDialKG</pwcdataset>
-      <video href="2021.emnlp-main.184.mp4"/>
     </paper>
     <paper id="185">
       <title><fixed-case>D</fixed-case>ialogue<fixed-case>CSE</fixed-case>: Dialogue-based Contrastive Learning of Sentence Embeddings</title>
@@ -2859,9 +2860,9 @@
       <url hash="eeb39f37">2021.emnlp-main.185</url>
       <bibkey>liu-etal-2021-dialoguecse</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.185</doi>
+      <video href="2021.emnlp-main.185.mp4"/>
       <pwccode url="https://github.com/wangruicn/dialoguecse" additional="false">wangruicn/dialoguecse</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.185.mp4"/>
     </paper>
     <paper id="186">
       <title>Improving Graph-based Sentence Ordering with Iteratively Predicted Pairwise Orderings</title>
@@ -2879,8 +2880,8 @@
       <url hash="01af5774">2021.emnlp-main.186</url>
       <bibkey>lai-etal-2021-improving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.186</doi>
-      <pwccode url="https://github.com/deeplearnxmu/irseg" additional="false">deeplearnxmu/irseg</pwccode>
       <video href="2021.emnlp-main.186.mp4"/>
+      <pwccode url="https://github.com/deeplearnxmu/irseg" additional="false">deeplearnxmu/irseg</pwccode>
     </paper>
     <paper id="187">
       <title>Not Just Classification: Recognizing Implicit Discourse Relation on Joint Modeling of Classification and Generation</title>
@@ -2920,10 +2921,10 @@
       <attachment type="Software" hash="4757fc0f">2021.emnlp-main.189.Software.zip</attachment>
       <bibkey>cheng-etal-2021-multimodal</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.189</doi>
+      <video href="2021.emnlp-main.189.mp4"/>
       <pwccode url="https://github.com/chengjunyan1/sp-transformer" additional="false">chengjunyan1/sp-transformer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multimodal-opinionlevel-sentiment-intensity">Multimodal Opinionlevel Sentiment Intensity</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ur-funny">UR-FUNNY</pwcdataset>
-      <video href="2021.emnlp-main.189.mp4"/>
     </paper>
     <paper id="190">
       <title>Hierarchical Multi-label Text Classification with Horizontal and Vertical Category Correlations</title>
@@ -2940,8 +2941,8 @@
       <attachment type="Software" hash="5291a014">2021.emnlp-main.190.Software.zip</attachment>
       <bibkey>xu-etal-2021-hierarchical</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.190</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/web-of-science-dataset">WOS</pwcdataset>
       <video href="2021.emnlp-main.190.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/web-of-science-dataset">WOS</pwcdataset>
     </paper>
     <paper id="191">
       <title><fixed-case>R</fixed-case>ank<fixed-case>NAS</fixed-case>: Efficient Neural Architecture Search by Pairwise Ranking</title>
@@ -2959,6 +2960,7 @@
       <bibkey>hu-etal-2021-ranknas</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.191</doi>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="192">
       <title><fixed-case>FL</fixed-case>i<fixed-case>T</fixed-case>ext: A Faster and Lighter Semi-Supervised Text Classification with Convolution Networks</title>
@@ -2972,9 +2974,9 @@
       <url hash="965d036a">2021.emnlp-main.192</url>
       <bibkey>liu-etal-2021-flitext</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.192</doi>
+      <video href="2021.emnlp-main.192.mp4"/>
       <pwccode url="https://github.com/valuesimplex/flitext" additional="false">valuesimplex/flitext</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.192.mp4"/>
     </paper>
     <paper id="193">
       <title>Evaluating Debiasing Techniques for Intersectional Biases</title>
@@ -3016,9 +3018,9 @@
       <attachment type="Software" hash="f7d33825">2021.emnlp-main.195.Software.zip</attachment>
       <bibkey>xiao-etal-2021-transductive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.195</doi>
+      <video href="2021.emnlp-main.195.mp4"/>
       <pwccode url="https://github.com/xiaofei05/tsst" additional="false">xiaofei05/tsst</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
-      <video href="2021.emnlp-main.195.mp4"/>
     </paper>
     <paper id="196">
       <title>Integrating Semantic Scenario and Word Relations for Abstractive Sentence Summarization</title>
@@ -3044,9 +3046,9 @@
       <url hash="fc13cfe4">2021.emnlp-main.197</url>
       <bibkey>tan-etal-2021-coupling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.197</doi>
+      <video href="2021.emnlp-main.197.mp4"/>
       <pwccode url="https://github.com/txannie/zp-dnlg" additional="false">txannie/zp-dnlg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/matinf">MATINF</pwcdataset>
-      <video href="2021.emnlp-main.197.mp4"/>
     </paper>
     <paper id="198">
       <title>Adaptive Bridge between Training and Inference for Dialogue Generation</title>
@@ -3079,8 +3081,8 @@
       <url hash="f4b77498">2021.emnlp-main.199</url>
       <bibkey>meng-etal-2021-conrpg</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.199</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <video href="2021.emnlp-main.199.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
     </paper>
     <paper id="200">
       <title>Building the Directed Semantic Graph for Coherent Long Text Generation</title>
@@ -3104,10 +3106,10 @@
       <url hash="e6ad3706">2021.emnlp-main.201</url>
       <bibkey>fei-etal-2021-iterative</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.201</doi>
+      <video href="2021.emnlp-main.201.mp4"/>
       <pwccode url="https://github.com/sion-zcfei/ignd" additional="false">sion-zcfei/ignd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.201.mp4"/>
     </paper>
     <paper id="202">
       <title>Asking Questions Like Educational Experts: <fixed-case>A</fixed-case>utomatically Generating Question-Answer Pairs on Real-World Examination Data</title>
@@ -3119,9 +3121,9 @@
       <url hash="edcc18b7">2021.emnlp-main.202</url>
       <bibkey>qu-etal-2021-asking</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.202</doi>
+      <video href="2021.emnlp-main.202.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.202.mp4"/>
     </paper>
     <paper id="203">
       <title>Syntactically-Informed Unsupervised Paraphrasing with Non-Parallel Data</title>
@@ -3138,9 +3140,9 @@
       <url hash="3b80fcdd">2021.emnlp-main.203</url>
       <bibkey>yang-etal-2021-syntactically</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.203</doi>
+      <video href="2021.emnlp-main.203.mp4"/>
       <pwccode url="https://github.com/lanse-sir/sup" additional="false">lanse-sir/sup</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.203.mp4"/>
     </paper>
     <paper id="204">
       <title>Exploring Task Difficulty for Few-Shot Relation Extraction</title>
@@ -3152,10 +3154,10 @@
       <url hash="a582ceff">2021.emnlp-main.204</url>
       <bibkey>han-etal-2021-exploring</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.204</doi>
+      <video href="2021.emnlp-main.204.mp4"/>
       <pwccode url="https://github.com/hanjiale/hcrp" additional="false">hanjiale/hcrp</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel-2-0">FewRel 2.0</pwcdataset>
-      <video href="2021.emnlp-main.204.mp4"/>
     </paper>
     <paper id="205">
       <title><fixed-case>M</fixed-case>u<fixed-case>VER</fixed-case>: <fixed-case>I</fixed-case>mproving First-Stage Entity Retrieval with Multi-View Entity Representations</title>
@@ -3171,9 +3173,9 @@
       <url hash="e45cddaa">2021.emnlp-main.205</url>
       <bibkey>ma-etal-2021-muver</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.205</doi>
+      <video href="2021.emnlp-main.205.mp4"/>
       <pwccode url="https://github.com/alibaba-nlp/muver" additional="false">alibaba-nlp/muver</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/zeshel">ZESHEL</pwcdataset>
-      <video href="2021.emnlp-main.205.mp4"/>
     </paper>
     <paper id="206">
       <title>Treasures Outside Contexts: Improving Event Detection via Global Statistics</title>
@@ -3186,8 +3188,8 @@
       <url hash="e908f3d5">2021.emnlp-main.206</url>
       <bibkey>li-etal-2021-treasures</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.206</doi>
-      <pwccode url="https://github.com/buted/ssjdn" additional="false">buted/ssjdn</pwccode>
       <video href="2021.emnlp-main.206.mp4"/>
+      <pwccode url="https://github.com/buted/ssjdn" additional="false">buted/ssjdn</pwccode>
     </paper>
     <paper id="207">
       <title>Uncertain Local-to-Global Networks for Document-Level Event Factuality Identification</title>
@@ -3201,8 +3203,8 @@
       <url hash="dbc7619b">2021.emnlp-main.207</url>
       <bibkey>cao-etal-2021-uncertain</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.207</doi>
-      <pwccode url="https://github.com/cpf-nlpr/ulgn4docefi" additional="false">cpf-nlpr/ulgn4docefi</pwccode>
       <video href="2021.emnlp-main.207.mp4"/>
+      <pwccode url="https://github.com/cpf-nlpr/ulgn4docefi" additional="false">cpf-nlpr/ulgn4docefi</pwccode>
     </paper>
     <paper id="208">
       <title>A Novel Global Feature-Oriented Relational Triple Extraction Model based on Table Filling</title>
@@ -3218,9 +3220,9 @@
       <url hash="48df7407">2021.emnlp-main.208</url>
       <bibkey>ren-etal-2021-novel</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.208</doi>
+      <video href="2021.emnlp-main.208.mp4"/>
       <pwccode url="https://github.com/neukg/GRTE" additional="false">neukg/GRTE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.emnlp-main.208.mp4"/>
     </paper>
     <paper id="209">
       <title>Structure-Augmented Keyphrase Generation</title>
@@ -3233,9 +3235,9 @@
       <url hash="2d5c9e7e">2021.emnlp-main.209</url>
       <bibkey>kim-etal-2021-structure</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.209</doi>
+      <video href="2021.emnlp-main.209.mp4"/>
       <pwccode url="https://github.com/jihyukkim-nlp/straugkg" additional="false">jihyukkim-nlp/straugkg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
-      <video href="2021.emnlp-main.209.mp4"/>
     </paper>
     <paper id="210">
       <title>An Empirical Study on Multiple Information Sources for Zero-Shot Fine-Grained Entity Typing</title>
@@ -3282,9 +3284,9 @@
       <url hash="13bb9bc0">2021.emnlp-main.212</url>
       <bibkey>dong-etal-2021-mapre</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.212</doi>
+      <video href="2021.emnlp-main.212.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.212.mp4"/>
     </paper>
     <paper id="213">
       <title>Heterogeneous Graph Neural Networks for Keyphrase Generation</title>
@@ -3297,9 +3299,9 @@
       <url hash="853fd24b">2021.emnlp-main.213</url>
       <bibkey>ye-etal-2021-heterogeneous</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.213</doi>
+      <video href="2021.emnlp-main.213.mp4"/>
       <pwccode url="https://github.com/jiacheng-ye/kg_gater" additional="false">jiacheng-ye/kg_gater</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
-      <video href="2021.emnlp-main.213.mp4"/>
     </paper>
     <paper id="214">
       <title>Machine Reading Comprehension as Data Augmentation: A Case Study on Implicit Event Argument Extraction</title>
@@ -3313,8 +3315,8 @@
       <doi>10.18653/v1/2021.emnlp-main.214</doi>
       <revision id="1" href="2021.emnlp-main.214v1" hash="981943cb"/>
       <revision id="2" href="2021.emnlp-main.214v2" hash="7a22ed6d" date="2022-01-02">Corrected sponsor number in the Acknowledgments section.</revision>
-      <pwcdataset url="https://paperswithcode.com/dataset/wikievents">WikiEvents</pwcdataset>
       <video href="2021.emnlp-main.214.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikievents">WikiEvents</pwcdataset>
     </paper>
     <paper id="215">
       <title><fixed-case>I</fixed-case>mportance <fixed-case>E</fixed-case>stimation from <fixed-case>M</fixed-case>ultiple <fixed-case>P</fixed-case>erspectives for <fixed-case>K</fixed-case>eyphrase <fixed-case>E</fixed-case>xtraction</title>
@@ -3326,8 +3328,8 @@
       <url hash="134adfb5">2021.emnlp-main.215</url>
       <bibkey>song-etal-2021-importance</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.215</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
       <video href="2021.emnlp-main.215.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
     </paper>
     <paper id="216">
       <title>Gradient Imitation Reinforcement Learning for Low Resource Relation Extraction</title>
@@ -3343,9 +3345,9 @@
       <url hash="8a08b757">2021.emnlp-main.216</url>
       <bibkey>hu-etal-2021-gradient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.216</doi>
+      <video href="2021.emnlp-main.216.mp4"/>
       <pwccode url="https://github.com/thu-bpm/gradlre" additional="false">thu-bpm/gradlre</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/semeval-2010-task-8">SemEval-2010 Task 8</pwcdataset>
-      <video href="2021.emnlp-main.216.mp4"/>
     </paper>
     <paper id="217">
       <title>Low-resource Taxonomy Enrichment with Pretrained Language Models</title>
@@ -3372,8 +3374,8 @@
       <url hash="b149fbb2">2021.emnlp-main.218</url>
       <bibkey>zhang-etal-2021-entity</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.218</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/funsd">FUNSD</pwcdataset>
       <video href="2021.emnlp-main.218.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/funsd">FUNSD</pwcdataset>
     </paper>
     <paper id="219">
       <title>Synchronous Dual Network with Cross-Type Attention for Joint Entity and Relation Extraction</title>
@@ -3384,8 +3386,8 @@
       <url hash="196e4033">2021.emnlp-main.219</url>
       <bibkey>wu-shi-2021-synchronous</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.219</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
       <video href="2021.emnlp-main.219.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
     </paper>
     <paper id="220">
       <title>Less is More: Pretrain a Strong <fixed-case>S</fixed-case>iamese Encoder for Dense Text Retrieval Using a Weak Decoder</title>
@@ -3403,12 +3405,12 @@
       <url hash="dcdc6015">2021.emnlp-main.220</url>
       <bibkey>lu-etal-2021-less</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.220</doi>
+      <video href="2021.emnlp-main.220.mp4"/>
       <pwccode url="https://github.com/microsoft/seed-encoder" additional="false">microsoft/seed-encoder</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.emnlp-main.220.mp4"/>
     </paper>
     <paper id="221">
       <title><fixed-case>T</fixed-case>rans<fixed-case>P</fixed-case>rompt: Towards an Automatic Transferable Prompting Framework for Few-shot Text Classification</title>
@@ -3422,12 +3424,12 @@
       <url hash="a897e977">2021.emnlp-main.221</url>
       <bibkey>wang-etal-2021-transprompt</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.221</doi>
+      <video href="2021.emnlp-main.221.mp4"/>
       <pwccode url="https://github.com/wjn1996/TransPrompt" additional="true">wjn1996/TransPrompt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mr">MR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.221.mp4"/>
     </paper>
     <paper id="222">
       <title>Weakly-supervised Text Classification Based on Keyword Graph</title>
@@ -3441,10 +3443,10 @@
       <url hash="fbf2cffe">2021.emnlp-main.222</url>
       <bibkey>zhang-etal-2021-weakly</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.222</doi>
+      <video href="2021.emnlp-main.222.mp4"/>
       <pwccode url="https://github.com/zhanglu-cst/classkg" additional="false">zhanglu-cst/classkg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.222.mp4"/>
     </paper>
     <paper id="223">
       <title>Efficient-<fixed-case>F</fixed-case>ed<fixed-case>R</fixed-case>ec: Efficient Federated Learning Framework for Privacy-Preserving News Recommendation</title>
@@ -3459,9 +3461,9 @@
       <url hash="5c3d6b96">2021.emnlp-main.223</url>
       <bibkey>yi-etal-2021-efficient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.223</doi>
+      <video href="2021.emnlp-main.223.mp4"/>
       <pwccode url="https://github.com/yjw1029/efficient-fedrec" additional="false">yjw1029/efficient-fedrec</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
-      <video href="2021.emnlp-main.223.mp4"/>
     </paper>
     <paper id="224">
       <title><fixed-case>R</fixed-case>ocket<fixed-case>QA</fixed-case>v2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking</title>
@@ -3478,10 +3480,10 @@
       <url hash="f214ef52">2021.emnlp-main.224</url>
       <bibkey>ren-etal-2021-rocketqav2</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.224</doi>
+      <video href="2021.emnlp-main.224.mp4"/>
       <pwccode url="https://github.com/paddlepaddle/rocketqa" additional="false">paddlepaddle/rocketqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.emnlp-main.224.mp4"/>
     </paper>
     <paper id="225">
       <title>Dealing with Typos for <fixed-case>BERT</fixed-case>-based Passage Retrieval and Ranking</title>
@@ -3492,9 +3494,9 @@
       <url hash="cfa52b4f">2021.emnlp-main.225</url>
       <bibkey>zhuang-zuccon-2021-dealing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.225</doi>
+      <video href="2021.emnlp-main.225.mp4"/>
       <pwccode url="https://github.com/ielab/typos-aware-bert" additional="true">ielab/typos-aware-bert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
-      <video href="2021.emnlp-main.225.mp4"/>
     </paper>
     <paper id="226">
       <title>From Alignment to Assignment: Frustratingly Simple Unsupervised Entity Alignment</title>
@@ -3508,9 +3510,9 @@
       <attachment type="Software" hash="d5f38a73">2021.emnlp-main.226.Software.zip</attachment>
       <bibkey>mao-etal-2021-alignment</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.226</doi>
+      <video href="2021.emnlp-main.226.mp4"/>
       <pwccode url="https://github.com/maoxinn/seu" additional="false">maoxinn/seu</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dbp15k">DBP15K</pwcdataset>
-      <video href="2021.emnlp-main.226.mp4"/>
     </paper>
     <paper id="227">
       <title>Simple and Effective Unsupervised Redundancy Elimination to Compress Dense Vectors for Passage Retrieval</title>
@@ -3536,8 +3538,8 @@
       <url hash="91258c4e">2021.emnlp-main.228</url>
       <bibkey>qin-etal-2021-relation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.228</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/semeval-2010-task-8">SemEval-2010 Task 8</pwcdataset>
       <video href="2021.emnlp-main.228.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/semeval-2010-task-8">SemEval-2010 Task 8</pwcdataset>
     </paper>
     <paper id="229">
       <title>A <fixed-case>B</fixed-case>ayesian Framework for Information-Theoretic Probing</title>
@@ -3548,8 +3550,8 @@
       <url hash="5c30af52">2021.emnlp-main.229</url>
       <bibkey>pimentel-cotterell-2021-bayesian</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.229</doi>
-      <pwccode url="https://github.com/rycolab/bayesian-mi" additional="false">rycolab/bayesian-mi</pwccode>
       <video href="2021.emnlp-main.229.mp4"/>
+      <pwccode url="https://github.com/rycolab/bayesian-mi" additional="false">rycolab/bayesian-mi</pwccode>
     </paper>
     <paper id="230">
       <title>Masked Language Modeling and the Distributional Hypothesis: Order Word Matters Pre-training for Little</title>
@@ -3564,6 +3566,7 @@
       <url hash="3799ae01">2021.emnlp-main.230</url>
       <bibkey>sinha-etal-2021-masked</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.230</doi>
+      <video href="2021.emnlp-main.230.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
@@ -3571,7 +3574,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.emnlp-main.230.mp4"/>
     </paper>
     <paper id="231">
       <title>What’s Hidden in a One-layer Randomly Weighted Transformer?</title>
@@ -3585,9 +3587,9 @@
       <url hash="3a1fa07a">2021.emnlp-main.231</url>
       <bibkey>shen-etal-2021-whats</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.231</doi>
+      <video href="2021.emnlp-main.231.mp4"/>
       <pwccode url="https://github.com/sincerass/one_layer_lottery_ticket" additional="false">sincerass/one_layer_lottery_ticket</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
-      <video href="2021.emnlp-main.231.mp4"/>
     </paper>
     <paper id="232">
       <title>Rethinking Denoised Auto-Encoding in Language Pre-Training</title>
@@ -3603,10 +3605,10 @@
       <url hash="6dc6ac5b">2021.emnlp-main.232</url>
       <bibkey>luo-etal-2021-rethinking</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.232</doi>
+      <video href="2021.emnlp-main.232.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gqa">GQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.emnlp-main.232.mp4"/>
     </paper>
     <paper id="233">
       <title>Lifelong Explainer for Lifelong Learners</title>
@@ -3620,8 +3622,8 @@
       <url hash="db337028">2021.emnlp-main.233</url>
       <bibkey>situ-etal-2021-lifelong</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.233</doi>
-      <pwccode url="https://github.com/situsnow/lle" additional="false">situsnow/lle</pwccode>
       <video href="2021.emnlp-main.233.mp4"/>
+      <pwccode url="https://github.com/situsnow/lle" additional="false">situsnow/lle</pwccode>
     </paper>
     <paper id="234">
       <title>Linguistic Dependencies and Statistical Dependence</title>
@@ -3634,10 +3636,10 @@
       <url hash="48121ef9">2021.emnlp-main.234</url>
       <bibkey>hoover-etal-2021-linguistic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.234</doi>
+      <video href="2021.emnlp-main.234.mp4"/>
       <pwccode url="https://github.com/mcqll/cpmi-dependencies" additional="false">mcqll/cpmi-dependencies</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.emnlp-main.234.mp4"/>
     </paper>
     <paper id="235">
       <title>Modeling Human Sentence Processing with Left-Corner Recurrent Neural Network Grammars</title>
@@ -3664,10 +3666,10 @@
       <url hash="3216324f">2021.emnlp-main.236</url>
       <bibkey>chen-etal-2021-simple</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.236</doi>
+      <video href="2021.emnlp-main.236.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xtreme">XTREME</pwcdataset>
-      <video href="2021.emnlp-main.236.mp4"/>
     </paper>
     <paper id="237">
       <title>Explore Better Relative Position Embeddings from Encoding Perspective for Transformer Models</title>
@@ -3679,6 +3681,7 @@
       <url hash="9bc84909">2021.emnlp-main.237</url>
       <bibkey>qu-etal-2021-explore</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.237</doi>
+      <video href="2021.emnlp-main.237.mp4"/>
       <pwccode url="https://github.com/menghuanlater/lfhc-gcdf-rpe" additional="false">menghuanlater/lfhc-gcdf-rpe</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cmrc">CMRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
@@ -3686,7 +3689,7 @@
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.emnlp-main.237.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="238">
       <title>Adversarial Mixing Policy for Relaxing Locally Linear Constraints in Mixup</title>
@@ -3700,9 +3703,9 @@
       <url hash="d239d82a">2021.emnlp-main.238</url>
       <bibkey>liu-etal-2021-adversarial</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.238</doi>
+      <video href="2021.emnlp-main.238.mp4"/>
       <pwccode url="https://github.com/pai-smallisallyourneed/mixup-amp" additional="false">pai-smallisallyourneed/mixup-amp</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.238.mp4"/>
     </paper>
     <paper id="239">
       <title>Is this the end of the gold standard? A straightforward reference-less grammatical error correction metric</title>
@@ -3769,6 +3772,7 @@
       <url hash="b7481d5c">2021.emnlp-main.243</url>
       <bibkey>lester-etal-2021-power</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.243</doi>
+      <video href="2021.emnlp-main.243.mp4"/>
       <pwccode url="https://github.com/google-research/prompt-tuning" additional="true">google-research/prompt-tuning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
@@ -3783,7 +3787,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
-      <video href="2021.emnlp-main.243.mp4"/>
     </paper>
     <paper id="244">
       <title>Scalable Font Reconstruction with Dual Latent Manifolds</title>
@@ -3811,8 +3814,8 @@
       <url hash="028d470d">2021.emnlp-main.245</url>
       <bibkey>chaudhury-etal-2021-neuro</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.245</doi>
-      <pwccode url="https://github.com/subhajit1411/slate-text-based-rl" additional="false">subhajit1411/slate-text-based-rl</pwccode>
       <video href="2021.emnlp-main.245.mp4"/>
+      <pwccode url="https://github.com/subhajit1411/slate-text-based-rl" additional="false">subhajit1411/slate-text-based-rl</pwccode>
     </paper>
     <paper id="246">
       <title>Layer-wise Model Pruning based on Mutual Information</title>
@@ -3828,9 +3831,9 @@
       <url hash="8aea531e">2021.emnlp-main.246</url>
       <bibkey>fan-etal-2021-layer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.246</doi>
+      <video href="2021.emnlp-main.246.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.246.mp4"/>
     </paper>
     <paper id="247">
       <title>Hierarchical Heterogeneous Graph Representation Learning for Short Text Classification</title>
@@ -3843,8 +3846,8 @@
       <url hash="9ac55a8c">2021.emnlp-main.247</url>
       <bibkey>wang-etal-2021-hierarchical</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.247</doi>
-      <pwccode url="https://github.com/tata1661/shine-emnlp21" additional="false">tata1661/shine-emnlp21</pwccode>
       <video href="2021.emnlp-main.247.mp4"/>
+      <pwccode url="https://github.com/tata1661/shine-emnlp21" additional="false">tata1661/shine-emnlp21</pwccode>
     </paper>
     <paper id="248">
       <title><tex-math>k</tex-math><fixed-case>F</fixed-case>olden: <tex-math>k</tex-math>-Fold Ensemble for Out-Of-Distribution Detection</title>
@@ -3875,10 +3878,10 @@
       <url hash="113f5d0c">2021.emnlp-main.249</url>
       <bibkey>yamaguchi-etal-2021-frustratingly</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.249</doi>
+      <video href="2021.emnlp-main.249.mp4"/>
       <pwccode url="https://github.com/gucci-j/light-transformer-emnlp2021" additional="false">gucci-j/light-transformer-emnlp2021</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.249.mp4"/>
     </paper>
     <paper id="250">
       <title><fixed-case>HRKD</fixed-case>: Hierarchical Relational Knowledge Distillation for Cross-domain Language Model Compression</title>
@@ -3891,8 +3894,8 @@
       <url hash="4c387f4d">2021.emnlp-main.250</url>
       <bibkey>dong-etal-2021-hrkd</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.250</doi>
-      <pwccode url="https://github.com/cheneydon/hrkd" additional="false">cheneydon/hrkd</pwccode>
       <video href="2021.emnlp-main.250.mp4"/>
+      <pwccode url="https://github.com/cheneydon/hrkd" additional="false">cheneydon/hrkd</pwccode>
     </paper>
     <paper id="251">
       <title>Searching for an Effective Defender: Benchmarking Defense against Adversarial Word Substitution</title>
@@ -3909,10 +3912,10 @@
       <url hash="54279453">2021.emnlp-main.251</url>
       <bibkey>li-etal-2021-searching</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.251</doi>
+      <video href="2021.emnlp-main.251.mp4"/>
       <pwccode url="https://github.com/rockylzy/textdefender" additional="false">rockylzy/textdefender</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.251.mp4"/>
     </paper>
     <paper id="252">
       <title>Re-embedding Difficult Samples via Mutual Information Constrained Semantically Oversampling for Imbalanced Text Classification</title>
@@ -3983,9 +3986,9 @@
       <url hash="5b6c5354">2021.emnlp-main.256</url>
       <bibkey>wang-etal-2021-neural</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.256</doi>
+      <video href="2021.emnlp-main.256.mp4"/>
       <pwccode url="https://github.com/Vicky-Wil/topic-NMT" additional="false">Vicky-Wil/topic-NMT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
-      <video href="2021.emnlp-main.256.mp4"/>
     </paper>
     <paper id="257">
       <title>Allocating Large Vocabulary Capacity for Cross-Lingual Language Model Pre-Training</title>
@@ -4002,13 +4005,13 @@
       <url hash="74d3e7fe">2021.emnlp-main.257</url>
       <bibkey>zheng-etal-2021-allocating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.257</doi>
+      <video href="2021.emnlp-main.257.mp4"/>
       <pwccode url="https://github.com/bozheng-hit/vocapxlm" additional="false">bozheng-hit/vocapxlm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.emnlp-main.257.mp4"/>
     </paper>
     <paper id="258">
       <title>Recurrent Attention for Neural Machine Translation</title>
@@ -4022,8 +4025,8 @@
       <url hash="698d050a">2021.emnlp-main.258</url>
       <bibkey>zeng-etal-2021-recurrent</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.258</doi>
-      <pwccode url="https://github.com/lemon0830/ran" additional="false">lemon0830/ran</pwccode>
       <video href="2021.emnlp-main.258.mp4"/>
+      <pwccode url="https://github.com/lemon0830/ran" additional="false">lemon0830/ran</pwccode>
     </paper>
     <paper id="259">
       <title>Learning from Multiple Noisy Augmented Data Sets for Better Cross-Lingual Spoken Language Understanding</title>
@@ -4065,8 +4068,8 @@
       <url hash="2b55a15a">2021.emnlp-main.261</url>
       <bibkey>li-etal-2021-unsupervised-neural</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.261</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <video href="2021.emnlp-main.261.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
     </paper>
     <paper id="262">
       <title>Encouraging Lexical Translation Consistency for Document-Level Neural Machine Translation</title>
@@ -4104,8 +4107,8 @@
       <url hash="36d6002e">2021.emnlp-main.264</url>
       <bibkey>liu-etal-2021-scheduled-sampling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.264</doi>
-      <pwccode url="https://github.com/adaxry/ss_on_decoding_steps" additional="false">adaxry/ss_on_decoding_steps</pwccode>
       <video href="2021.emnlp-main.264.mp4"/>
+      <pwccode url="https://github.com/adaxry/ss_on_decoding_steps" additional="false">adaxry/ss_on_decoding_steps</pwccode>
     </paper>
     <paper id="265">
       <title>Learning to Rewrite for Non-Autoregressive Neural Machine Translation</title>
@@ -4117,8 +4120,8 @@
       <url hash="398efbbb">2021.emnlp-main.265</url>
       <bibkey>geng-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.265</doi>
-      <pwccode url="https://github.com/xwgeng/rewritenat" additional="false">xwgeng/rewritenat</pwccode>
       <video href="2021.emnlp-main.265.mp4"/>
+      <pwccode url="https://github.com/xwgeng/rewritenat" additional="false">xwgeng/rewritenat</pwccode>
     </paper>
     <paper id="266">
       <title><fixed-case>SHAPE</fixed-case>: <fixed-case>S</fixed-case>hifted Absolute Position Embedding for Transformers</title>
@@ -4161,8 +4164,8 @@
       <url hash="c461d256">2021.emnlp-main.268</url>
       <bibkey>vu-etal-2021-generalised</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.268</doi>
-      <pwccode url="https://github.com/trangvu/guda" additional="false">trangvu/guda</pwccode>
       <video href="2021.emnlp-main.268.mp4"/>
+      <pwccode url="https://github.com/trangvu/guda" additional="false">trangvu/guda</pwccode>
     </paper>
     <paper id="269">
       <title><fixed-case>STANKER</fixed-case>: Stacking Network based on Level-grained Attention-masked <fixed-case>BERT</fixed-case> for Rumor Detection on Social Media</title>
@@ -4176,8 +4179,8 @@
       <attachment type="Software" hash="91e89325">2021.emnlp-main.269.Software.zip</attachment>
       <bibkey>rao-etal-2021-stanker</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.269</doi>
-      <pwccode url="https://github.com/fip-lab/stanker" additional="false">fip-lab/stanker</pwccode>
       <video href="2021.emnlp-main.269.mp4"/>
+      <pwccode url="https://github.com/fip-lab/stanker" additional="false">fip-lab/stanker</pwccode>
     </paper>
     <paper id="270">
       <title><fixed-case>A</fixed-case>ctive<fixed-case>EA</fixed-case>: Active Learning for Neural Entity Alignment</title>
@@ -4191,8 +4194,8 @@
       <url hash="5de60a10">2021.emnlp-main.270</url>
       <bibkey>liu-etal-2021-activeea</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.270</doi>
-      <pwccode url="https://github.com/uq-neusoft-health-data-science/activeea" additional="false">uq-neusoft-health-data-science/activeea</pwccode>
       <video href="2021.emnlp-main.270.mp4"/>
+      <pwccode url="https://github.com/uq-neusoft-health-data-science/activeea" additional="false">uq-neusoft-health-data-science/activeea</pwccode>
     </paper>
     <paper id="271">
       <title>Cost-effective End-to-end Information Extraction for Semi-structured Document Images</title>
@@ -4219,8 +4222,8 @@
       <url hash="877db22b">2021.emnlp-main.272</url>
       <bibkey>yu-etal-2021-improving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.272</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/math23k">Math23K</pwcdataset>
       <video href="2021.emnlp-main.272.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/math23k">Math23K</pwcdataset>
     </paper>
     <paper id="273">
       <title><fixed-case>G</fixed-case>raph<fixed-case>MR</fixed-case>: Graph Neural Network for Mathematical Reasoning</title>
@@ -4280,11 +4283,11 @@
       <url hash="7d38683d">2021.emnlp-main.274</url>
       <bibkey>kim-etal-2021-changes</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.274</doi>
+      <video href="2021.emnlp-main.274.mp4"/>
       <pwccode url="" additional="true"/>
       <pwcdataset url="https://paperswithcode.com/dataset/klue">KLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/korquad">KorQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.274.mp4"/>
     </paper>
     <paper id="275">
       <title><fixed-case>APIR</fixed-case>ec<fixed-case>X</fixed-case>: Cross-Library <fixed-case>API</fixed-case> Recommendation via Pre-Trained Language Model</title>
@@ -4314,8 +4317,8 @@
       <url hash="6efbf08b">2021.emnlp-main.276</url>
       <bibkey>zhang-etal-2021-gmh</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.276</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/umls">UMLS</pwcdataset>
       <video href="2021.emnlp-main.276.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/umls">UMLS</pwcdataset>
     </paper>
     <paper id="277">
       <title><fixed-case>BPM</fixed-case>_<fixed-case>MT</fixed-case>: Enhanced Backchannel Prediction Model using Multi-Task Learning</title>
@@ -4344,9 +4347,9 @@
       <url hash="1ba7fce2">2021.emnlp-main.278</url>
       <bibkey>liu-etal-2021-graphine</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.278</doi>
+      <video href="2021.emnlp-main.278.mp4"/>
       <pwccode url="https://github.com/zequnl/graphex" additional="false">zequnl/graphex</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/graphine">Graphine</pwcdataset>
-      <video href="2021.emnlp-main.278.mp4"/>
     </paper>
     <paper id="279">
       <title>Leveraging Order-Free Tag Relations for Context-Aware Recommendation</title>
@@ -4436,8 +4439,8 @@
       <url hash="8389c8ba">2021.emnlp-main.284</url>
       <bibkey>yan-etal-2021-biomedical</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.284</doi>
-      <pwccode url="https://github.com/yan-cheng/bcnh" additional="false">yan-cheng/bcnh</pwccode>
       <video href="2021.emnlp-main.284.mp4"/>
+      <pwccode url="https://github.com/yan-cheng/bcnh" additional="false">yan-cheng/bcnh</pwccode>
     </paper>
     <paper id="285">
       <title>Leveraging Capsule Routing to Associate Knowledge with Medical Literature Hierarchically</title>
@@ -4453,8 +4456,8 @@
       <url hash="25fdec70">2021.emnlp-main.285</url>
       <bibkey>liu-etal-2021-leveraging</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.285</doi>
-      <pwccode url="https://github.com/gdls/hicapsrkl" additional="false">gdls/hicapsrkl</pwccode>
       <video href="2021.emnlp-main.285.mp4"/>
+      <pwccode url="https://github.com/gdls/hicapsrkl" additional="false">gdls/hicapsrkl</pwccode>
     </paper>
     <paper id="286">
       <title>Label-Enhanced Hierarchical Contextualized Representation for Sequential Metaphor Identification</title>
@@ -4499,9 +4502,9 @@
       <url hash="ff258315">2021.emnlp-main.288</url>
       <bibkey>nguyen-etal-2021-automated</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.288</doi>
+      <video href="2021.emnlp-main.288.mp4"/>
       <pwccode url="https://github.com/ginobilinie/xray_report_generation" additional="false">ginobilinie/xray_report_generation</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/chexpert">CheXpert</pwcdataset>
-      <video href="2021.emnlp-main.288.mp4"/>
     </paper>
     <paper id="289">
       <title>Enhancing Document Ranking with Task-adaptive Training and Segmented Token Recovery Mechanism</title>
@@ -4516,8 +4519,8 @@
       <url hash="a5e4c211">2021.emnlp-main.289</url>
       <bibkey>sun-etal-2021-enhancing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.289</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <video href="2021.emnlp-main.289.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
     </paper>
     <paper id="290">
       <title>Abstract, Rationale, Stance: A Joint Model for Scientific Claim Verification</title>
@@ -4530,9 +4533,9 @@
       <url hash="dff4f8d8">2021.emnlp-main.290</url>
       <bibkey>zhang-etal-2021-abstract</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.290</doi>
+      <video href="2021.emnlp-main.290.mp4"/>
       <pwccode url="https://github.com/zhiweizhang97/arsjointmodel" additional="false">zhiweizhang97/arsjointmodel</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
-      <video href="2021.emnlp-main.290.mp4"/>
     </paper>
     <paper id="291">
       <title>A Fine-Grained Domain Adaption Model for Joint Word Segmentation and <fixed-case>POS</fixed-case> Tagging</title>
@@ -4547,8 +4550,8 @@
       <url hash="aaa948cb">2021.emnlp-main.291</url>
       <bibkey>jiang-etal-2021-fine</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.291</doi>
-      <pwccode url="https://github.com/jzx555/fgda" additional="false">jzx555/fgda</pwccode>
       <video href="2021.emnlp-main.291.mp4"/>
+      <pwccode url="https://github.com/jzx555/fgda" additional="false">jzx555/fgda</pwccode>
     </paper>
     <paper id="292">
       <title>Answering Open-Domain Questions of Varying Reasoning Steps from Text</title>
@@ -4561,11 +4564,11 @@
       <url hash="e7506d3d">2021.emnlp-main.292</url>
       <bibkey>qi-etal-2021-answering</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.292</doi>
+      <video href="2021.emnlp-main.292.mp4"/>
       <pwccode url="https://github.com/beerqa/irrr" additional="false">beerqa/irrr</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.292.mp4"/>
     </paper>
     <paper id="293">
       <title>Adaptive Information Seeking for Open-Domain Question Answering</title>
@@ -4580,10 +4583,10 @@
       <attachment type="Software" hash="6bf9d68d">2021.emnlp-main.293.Software.zip</attachment>
       <bibkey>zhu-etal-2021-adaptive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.293</doi>
+      <video href="2021.emnlp-main.293.mp4"/>
       <pwccode url="https://github.com/zycdev/aiso" additional="false">zycdev/aiso</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.293.mp4"/>
     </paper>
     <paper id="294">
       <title>Mapping probability word problems to executable representations</title>
@@ -4599,8 +4602,8 @@
       <url hash="a8caa0e1">2021.emnlp-main.294</url>
       <bibkey>suster-etal-2021-mapping</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.294</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/mathqa">MathQA</pwcdataset>
       <video href="2021.emnlp-main.294.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mathqa">MathQA</pwcdataset>
     </paper>
     <paper id="295">
       <title>Enhancing Multiple-choice Machine Reading Comprehension by Punishing Illogical Interpretations</title>
@@ -4617,11 +4620,11 @@
       <url hash="e6b9cc09">2021.emnlp-main.295</url>
       <bibkey>ju-etal-2021-enhancing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.295</doi>
+      <video href="2021.emnlp-main.295.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dream">DREAM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multirc">MultiRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.emnlp-main.295.mp4"/>
     </paper>
     <paper id="296">
       <title>Large-Scale Relation Learning for Question Answering over Knowledge Bases with Pre-trained Language Models</title>
@@ -4638,9 +4641,9 @@
       <url hash="6bc5ec1b">2021.emnlp-main.296</url>
       <bibkey>yan-etal-2021-large</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.296</doi>
+      <video href="2021.emnlp-main.296.mp4"/>
       <pwccode url="https://github.com/yym6472/kbqarelationlearning" additional="false">yym6472/kbqarelationlearning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
-      <video href="2021.emnlp-main.296.mp4"/>
     </paper>
     <paper id="297">
       <title>Phrase Retrieval Learns Passage Retrieval, Too</title>
@@ -4652,12 +4655,12 @@
       <url hash="19074be1">2021.emnlp-main.297</url>
       <bibkey>lee-etal-2021-phrase</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.297</doi>
+      <video href="2021.emnlp-main.297.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/DensePhrases" additional="false">princeton-nlp/DensePhrases</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.emnlp-main.297.mp4"/>
     </paper>
     <paper id="298">
       <title>Neural Natural Logic Inference for Interpretable Question Answering</title>
@@ -4671,8 +4674,8 @@
       <url hash="494efaa8">2021.emnlp-main.298</url>
       <bibkey>shi-etal-2021-neural</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.298</doi>
-      <pwccode url="https://github.com/shijihao/neunli" additional="false">shijihao/neunli</pwccode>
       <video href="2021.emnlp-main.298.mp4"/>
+      <pwccode url="https://github.com/shijihao/neunli" additional="false">shijihao/neunli</pwccode>
     </paper>
     <paper id="299">
       <title>Smoothing Dialogue States for Open Conversational Machine Reading</title>
@@ -4686,9 +4689,9 @@
       <url hash="28e9ba53">2021.emnlp-main.299</url>
       <bibkey>zhang-etal-2021-smoothing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.299</doi>
+      <video href="2021.emnlp-main.299.mp4"/>
       <pwccode url="https://github.com/ozyyshr/oscar" additional="false">ozyyshr/oscar</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sharc">ShARC</pwcdataset>
-      <video href="2021.emnlp-main.299.mp4"/>
     </paper>
     <paper id="300">
       <title><fixed-case>F</fixed-case>in<fixed-case>QA</fixed-case>: A Dataset of Numerical Reasoning over Financial Data</title>
@@ -4708,11 +4711,11 @@
       <url hash="c58cbf87">2021.emnlp-main.300</url>
       <bibkey>chen-etal-2021-finqa</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.300</doi>
+      <video href="2021.emnlp-main.300.mp4"/>
       <pwccode url="https://github.com/czyssrs/finqa" additional="false">czyssrs/finqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/finqa">FinQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mathqa">MathQA</pwcdataset>
-      <video href="2021.emnlp-main.300.mp4"/>
     </paper>
     <paper id="301">
       <title><fixed-case>F</fixed-case>i<fixed-case>D</fixed-case>-Ex: Improving Sequence-to-Sequence Models for Extractive Rationale Generation</title>
@@ -4741,10 +4744,10 @@
       <url hash="74dd8e81">2021.emnlp-main.302</url>
       <bibkey>lin-etal-2021-rockner</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.302</doi>
+      <video href="2021.emnlp-main.302.mp4"/>
       <pwccode url="https://github.com/INK-USC/RockNER" additional="false">INK-USC/RockNER</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ontorock">OntoRock</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ontonotes-5-0">OntoNotes 5.0</pwcdataset>
-      <video href="2021.emnlp-main.302.mp4"/>
     </paper>
     <paper id="303">
       <title>Diagnosing the First-Order Logical Reasoning Ability Through <fixed-case>L</fixed-case>ogic<fixed-case>NLI</fixed-case></title>
@@ -4759,9 +4762,9 @@
       <url hash="07c8e797">2021.emnlp-main.303</url>
       <bibkey>tian-etal-2021-diagnosing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.303</doi>
+      <video href="2021.emnlp-main.303.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/logiqa">LogiQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/reclor">ReClor</pwcdataset>
-      <video href="2021.emnlp-main.303.mp4"/>
     </paper>
     <paper id="304">
       <title>Constructing a Psychometric Testbed for Fair Natural Language Processing</title>
@@ -4776,8 +4779,8 @@
       <url hash="1c61ed10">2021.emnlp-main.304</url>
       <bibkey>abbasi-etal-2021-constructing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.304</doi>
-      <pwccode url="https://github.com/nd-hal/fair-psych-nlp" additional="false">nd-hal/fair-psych-nlp</pwccode>
       <video href="2021.emnlp-main.304.mp4"/>
+      <pwccode url="https://github.com/nd-hal/fair-psych-nlp" additional="false">nd-hal/fair-psych-nlp</pwccode>
     </paper>
     <paper id="305">
       <title><fixed-case>COUGH</fixed-case>: A Challenge Dataset and Models for <fixed-case>COVID</fixed-case>-19 <fixed-case>FAQ</fixed-case> Retrieval</title>
@@ -4791,9 +4794,9 @@
       <url hash="d2a403c3">2021.emnlp-main.305</url>
       <bibkey>zhang-etal-2021-cough</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.305</doi>
+      <video href="2021.emnlp-main.305.mp4"/>
       <pwccode url="https://github.com/sunlab-osu/covid-faq" additional="false">sunlab-osu/covid-faq</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cough">COUGH</pwcdataset>
-      <video href="2021.emnlp-main.305.mp4"/>
     </paper>
     <paper id="306">
       <title><fixed-case>C</fixed-case>hinese <fixed-case>WPLC</fixed-case>: A <fixed-case>C</fixed-case>hinese Dataset for Evaluating Pretrained Language Models on Word Prediction Given Long-Range Context</title>
@@ -4806,11 +4809,11 @@
       <url hash="53771b29">2021.emnlp-main.306</url>
       <bibkey>ge-etal-2021-chinese</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.306</doi>
+      <video href="2021.emnlp-main.306.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cbt">CBT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/lambada">LAMBADA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winogrande">WinoGrande</pwcdataset>
-      <video href="2021.emnlp-main.306.mp4"/>
     </paper>
     <paper id="307">
       <title><fixed-case>W</fixed-case>ino<fixed-case>L</fixed-case>ogic: <fixed-case>A</fixed-case> Zero-Shot Logic-based Diagnostic Dataset for <fixed-case>W</fixed-case>inograd <fixed-case>S</fixed-case>chema <fixed-case>C</fixed-case>hallenge</title>
@@ -4824,12 +4827,12 @@
       <attachment type="Software" hash="af7b7a6e">2021.emnlp-main.307.Software.zip</attachment>
       <bibkey>he-etal-2021-winologic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.307</doi>
+      <video href="2021.emnlp-main.307.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winogrande">WinoGrande</pwcdataset>
-      <video href="2021.emnlp-main.307.mp4"/>
     </paper>
     <paper id="308">
       <title>Pseudo Zero Pronoun Resolution Improves Zero Anaphora Resolution</title>
@@ -4843,8 +4846,8 @@
       <url hash="f9a643c6">2021.emnlp-main.308</url>
       <bibkey>konno-etal-2021-pseudo</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.308</doi>
-      <pwccode url="https://github.com/ryuto10/pzero-improves-zar" additional="false">ryuto10/pzero-improves-zar</pwccode>
       <video href="2021.emnlp-main.308.mp4"/>
+      <pwccode url="https://github.com/ryuto10/pzero-improves-zar" additional="false">ryuto10/pzero-improves-zar</pwccode>
     </paper>
     <paper id="309">
       <title>Aligning Cross-lingual Sentence Representations with Dual Momentum Contrast</title>
@@ -4856,8 +4859,8 @@
       <url hash="b5197fa8">2021.emnlp-main.309</url>
       <bibkey>wang-etal-2021-aligning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.309</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
       <video href="2021.emnlp-main.309.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
     </paper>
     <paper id="310">
       <title>Total Recall: a Customized Continual Learning Method for Neural Semantic Parsers</title>
@@ -4869,8 +4872,8 @@
       <url hash="6b0a4707">2021.emnlp-main.310</url>
       <bibkey>li-etal-2021-total</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.310</doi>
-      <pwccode url="https://github.com/zhuang-li/cl_nsp" additional="false">zhuang-li/cl_nsp</pwccode>
       <video href="2021.emnlp-main.310.mp4"/>
+      <pwccode url="https://github.com/zhuang-li/cl_nsp" additional="false">zhuang-li/cl_nsp</pwccode>
     </paper>
     <paper id="311">
       <title>Exophoric Pronoun Resolution in Dialogues with Topic Regularization</title>
@@ -4885,9 +4888,9 @@
       <url hash="3cd4dfc9">2021.emnlp-main.311</url>
       <bibkey>yu-etal-2021-exophoric</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.311</doi>
+      <video href="2021.emnlp-main.311.mp4"/>
       <pwccode url="https://github.com/hkust-knowcomp/exo-pcr" additional="false">hkust-knowcomp/exo-pcr</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/vispro">VisPro</pwcdataset>
-      <video href="2021.emnlp-main.311.mp4"/>
     </paper>
     <paper id="312">
       <title>Context-Aware Interaction Network for Question Matching</title>
@@ -4915,8 +4918,8 @@
       <url hash="56d6c4f0">2021.emnlp-main.313</url>
       <bibkey>liu-etal-2021-temp</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.313</doi>
-      <pwccode url="https://github.com/liu-zichen/TEMP" additional="false">liu-zichen/TEMP</pwccode>
       <video href="2021.emnlp-main.313.mp4"/>
+      <pwccode url="https://github.com/liu-zichen/TEMP" additional="false">liu-zichen/TEMP</pwccode>
     </paper>
     <paper id="314">
       <title>A Graph-Based Neural Model for End-to-End Frame Semantic Parsing</title>
@@ -4928,9 +4931,9 @@
       <url hash="c845a9eb">2021.emnlp-main.314</url>
       <bibkey>lin-etal-2021-graph</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.314</doi>
+      <video href="2021.emnlp-main.314.mp4"/>
       <pwccode url="https://github.com/ch4osmy7h/framenetparser" additional="false">ch4osmy7h/framenetparser</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
-      <video href="2021.emnlp-main.314.mp4"/>
     </paper>
     <paper id="315">
       <title>Virtual Data Augmentation: A Robust and General Framework for Fine-tuning Pre-trained Models</title>
@@ -4945,11 +4948,11 @@
       <url hash="3eab498f">2021.emnlp-main.315</url>
       <bibkey>zhou-etal-2021-virtual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.315</doi>
+      <video href="2021.emnlp-main.315.mp4"/>
       <pwccode url="https://github.com/rucaibox/vda" additional="false">rucaibox/vda</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.emnlp-main.315.mp4"/>
     </paper>
     <paper id="316">
       <title><fixed-case>CATE</fixed-case>: A Contrastive Pre-trained Model for Metaphor Detection with Semi-supervised Learning</title>
@@ -4977,8 +4980,8 @@
       <attachment type="Software" hash="7213fab1">2021.emnlp-main.317.Software.zip</attachment>
       <bibkey>zhou-etal-2021-closer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.317</doi>
-      <pwccode url="https://github.com/zyxnlp/aclt" additional="false">zyxnlp/aclt</pwccode>
       <video href="2021.emnlp-main.317.mp4"/>
+      <pwccode url="https://github.com/zyxnlp/aclt" additional="false">zyxnlp/aclt</pwccode>
     </paper>
     <paper id="318">
       <title>Seeking Common but Distinguishing Difference, A Joint Aspect-based Sentiment Analysis Model</title>
@@ -4994,9 +4997,9 @@
       <revision id="2" href="2021.emnlp-main.318v2" hash="1b80efdd" date="2021-11-18">Deleted funding sources in acknowledgment.</revision>
       <doi>10.18653/v1/2021.emnlp-main.318</doi>
       <revision id="3" href="2021.emnlp-main.318v3" hash="e0b30006" date="2022-02-09">Updated Sections 2.2, 2.3, 3.1, Figure 2, Table 1, and Table 8.</revision>
+      <video href="2021.emnlp-main.318.mp4"/>
       <pwccode url="https://github.com/betahj/pairabsa" additional="false">betahj/pairabsa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/aste-data-v2">ASTE-Data-V2</pwcdataset>
-      <video href="2021.emnlp-main.318.mp4"/>
     </paper>
     <paper id="319">
       <title>Argument Pair Extraction with Mutual Guidance and Inter-sentence Relation Graph</title>
@@ -5011,8 +5014,8 @@
       <url hash="50f124ad">2021.emnlp-main.319</url>
       <bibkey>bao-etal-2021-argument</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.319</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/rr">RR</pwcdataset>
       <video href="2021.emnlp-main.319.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/rr">RR</pwcdataset>
     </paper>
     <paper id="320">
       <title>Emotion Inference in Multi-Turn Conversations with Addressee-Aware Module and Ensemble Strategy</title>
@@ -5028,10 +5031,10 @@
       <url hash="b35a3a56">2021.emnlp-main.320</url>
       <bibkey>li-etal-2021-emotion</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.320</doi>
+      <video href="2021.emnlp-main.320.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/emorynlp">EmoryNLP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
-      <video href="2021.emnlp-main.320.mp4"/>
     </paper>
     <paper id="321">
       <title>Improving Federated Learning for Aspect-based Sentiment Analysis via Topic Memories</title>
@@ -5071,10 +5074,10 @@
       <url hash="c4cd7d64">2021.emnlp-main.323</url>
       <bibkey>li-etal-2021-ctal</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.323</doi>
+      <video href="2021.emnlp-main.323.mp4"/>
       <pwccode url="https://github.com/ydkwim/ctal" additional="false">ydkwim/ctal</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
-      <video href="2021.emnlp-main.323.mp4"/>
     </paper>
     <paper id="324">
       <title>Relation-aware Video Reading Comprehension for Temporal Language Grounding</title>
@@ -5088,11 +5091,11 @@
       <url hash="bcdfc274">2021.emnlp-main.324</url>
       <bibkey>gao-etal-2021-relation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.324</doi>
+      <video href="2021.emnlp-main.324.mp4"/>
       <pwccode url="https://github.com/Huntersxsx/RaNet" additional="false">Huntersxsx/RaNet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/activitynet-captions">ActivityNet Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/charades">Charades</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/charades-sta">Charades-STA</pwcdataset>
-      <video href="2021.emnlp-main.324.mp4"/>
     </paper>
     <paper id="325">
       <title>Mutual-Learning Improves End-to-End Speech Translation</title>
@@ -5105,8 +5108,8 @@
       <url hash="29196ee5">2021.emnlp-main.325</url>
       <bibkey>zhao-etal-2021-mutual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.325</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
       <video href="2021.emnlp-main.325.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
     </paper>
     <paper id="326">
       <title>Vision Guided Generative Pre-trained Language Models for Multimodal Abstractive Summarization</title>
@@ -5120,9 +5123,9 @@
       <attachment type="Software" hash="cf748908">2021.emnlp-main.326.Software.zip</attachment>
       <bibkey>yu-etal-2021-vision</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.326</doi>
+      <video href="2021.emnlp-main.326.mp4"/>
       <pwccode url="https://github.com/hltchkust/vg-gplms" additional="false">hltchkust/vg-gplms</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/how2">How2</pwcdataset>
-      <video href="2021.emnlp-main.326.mp4"/>
     </paper>
     <paper id="327">
       <title>Natural Language Video Localization with Learnable Moment Proposals</title>
@@ -5136,10 +5139,10 @@
       <url hash="e06416f4">2021.emnlp-main.327</url>
       <bibkey>xiao-etal-2021-natural</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.327</doi>
+      <video href="2021.emnlp-main.327.mp4"/>
       <pwccode url="https://github.com/xiaoneil/lpnet" additional="false">xiaoneil/lpnet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/activitynet-captions">ActivityNet Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/charades-sta">Charades-STA</pwcdataset>
-      <video href="2021.emnlp-main.327.mp4"/>
     </paper>
     <paper id="328">
       <title>Language-Aligned Waypoint (<fixed-case>LAW</fixed-case>) Supervision for Vision-and-Language Navigation in Continuous Environments</title>
@@ -5177,8 +5180,8 @@
       <url hash="463fd16e">2021.emnlp-main.330</url>
       <bibkey>kwon-etal-2021-considering</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.330</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <video href="2021.emnlp-main.330.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
     </paper>
     <paper id="331">
       <title>Frame Semantic-Enhanced Sentence Modeling for Sentence-level Extractive Text Summarization</title>
@@ -5208,8 +5211,8 @@
       <url hash="99a6e237">2021.emnlp-main.332</url>
       <bibkey>shi-etal-2021-cast</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.332</doi>
-      <pwccode url="https://github.com/DeepSoftwareAnalytics/CAST" additional="false">DeepSoftwareAnalytics/CAST</pwccode>
       <video href="2021.emnlp-main.332.mp4"/>
+      <pwccode url="https://github.com/DeepSoftwareAnalytics/CAST" additional="false">DeepSoftwareAnalytics/CAST</pwccode>
     </paper>
     <paper id="333">
       <title><fixed-case>S</fixed-case>g<fixed-case>S</fixed-case>um:Transforming Multi-document Summarization into Sub-graph Selection</title>
@@ -5224,9 +5227,9 @@
       <url hash="ac499aee">2021.emnlp-main.333</url>
       <bibkey>chen-etal-2021-sgsum</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.333</doi>
+      <video href="2021.emnlp-main.333.mp4"/>
       <pwccode url="https://github.com/PaddlePaddle/Research/tree/master/NLP/EMNLP2021-SgSum" additional="false">PaddlePaddle/Research</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multi-news">Multi-News</pwcdataset>
-      <video href="2021.emnlp-main.333.mp4"/>
     </paper>
     <paper id="334">
       <title>Event Graph based Sentence Fusion</title>
@@ -5239,8 +5242,8 @@
       <attachment type="Software" hash="8ecc98de">2021.emnlp-main.334.Software.zip</attachment>
       <bibkey>yuan-etal-2021-event</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.334</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/multi-news">Multi-News</pwcdataset>
       <video href="2021.emnlp-main.334.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/multi-news">Multi-News</pwcdataset>
     </paper>
     <paper id="335">
       <title>Transformer-based Lexically Constrained Headline Generation</title>
@@ -5256,9 +5259,9 @@
       <url hash="07446807">2021.emnlp-main.335</url>
       <bibkey>yamada-etal-2021-transformer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.335</doi>
+      <video href="2021.emnlp-main.335.mp4"/>
       <pwccode url="https://github.com/asahi-research/script-for-transformer-based-seq2bf" additional="false">asahi-research/script-for-transformer-based-seq2bf</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/jnc">JNC</pwcdataset>
-      <video href="2021.emnlp-main.335.mp4"/>
     </paper>
     <paper id="336">
       <title>Learn to Copy from the Copying History: Correlational Copy Network for Abstractive Summarization</title>
@@ -5274,9 +5277,9 @@
       <url hash="58a606c5">2021.emnlp-main.336</url>
       <bibkey>li-etal-2021-learn</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.336</doi>
+      <video href="2021.emnlp-main.336.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.emnlp-main.336.mp4"/>
     </paper>
     <paper id="337">
       <title>Gradient-Based Adversarial Factual Consistency Evaluation for Abstractive Summarization</title>
@@ -5289,8 +5292,8 @@
       <url hash="fef4caa7">2021.emnlp-main.337</url>
       <bibkey>zeng-etal-2021-gradient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.337</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <video href="2021.emnlp-main.337.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
     </paper>
     <paper id="338">
       <title>Word Reordering for Zero-shot Cross-lingual Structured Prediction</title>
@@ -5337,12 +5340,12 @@
       <url hash="6a59894c">2021.emnlp-main.340</url>
       <bibkey>lyu-etal-2021-improving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.340</doi>
+      <video href="2021.emnlp-main.340.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/duorc">DuoRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsqa">NewsQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.emnlp-main.340.mp4"/>
     </paper>
     <paper id="341">
       <title><fixed-case>T</fixed-case>ransfer<fixed-case>N</fixed-case>et: An Effective and Transparent Framework for Multi-hop Question Answering over Relation Graph</title>
@@ -5356,11 +5359,11 @@
       <url hash="1c9332f4">2021.emnlp-main.341</url>
       <bibkey>shi-etal-2021-transfernet</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.341</doi>
+      <video href="2021.emnlp-main.341.mp4"/>
       <pwccode url="https://github.com/shijx12/TransferNet" additional="false">shijx12/TransferNet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/metaqa">MetaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/simplequestions">SimpleQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikimovies">WikiMovies</pwcdataset>
-      <video href="2021.emnlp-main.341.mp4"/>
     </paper>
     <paper id="342">
       <title>Topic Transferable Table Question Answering</title>
@@ -5377,10 +5380,10 @@
       <url hash="2c1cf0d2">2021.emnlp-main.342</url>
       <bibkey>chemmengath-etal-2021-topic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.342</doi>
+      <video href="2021.emnlp-main.342.mp4"/>
       <pwccode url="https://github.com/ibm/t3qa" additional="false">ibm/t3qa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitablequestions">WikiTableQuestions</pwcdataset>
-      <video href="2021.emnlp-main.342.mp4"/>
     </paper>
     <paper id="343">
       <title><fixed-case>W</fixed-case>eb<fixed-case>SRC</fixed-case>: A Dataset for Web-Based Structural Reading Comprehension</title>
@@ -5397,8 +5400,8 @@
       <url hash="12e79825">2021.emnlp-main.343</url>
       <bibkey>chen-etal-2021-websrc</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.343</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <video href="2021.emnlp-main.343.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
     </paper>
     <paper id="344">
       <title>Cryptonite: A Cryptic Crossword Benchmark for Extreme Ambiguity in Language</title>
@@ -5411,8 +5414,8 @@
       <url hash="79aef513">2021.emnlp-main.344</url>
       <bibkey>efrat-etal-2021-cryptonite</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.344</doi>
-      <pwccode url="https://github.com/aviaefrat/cryptonite" additional="false">aviaefrat/cryptonite</pwccode>
       <video href="2021.emnlp-main.344.mp4"/>
+      <pwccode url="https://github.com/aviaefrat/cryptonite" additional="false">aviaefrat/cryptonite</pwccode>
     </paper>
     <paper id="345">
       <title>End-to-End Entity Resolution and Question Answering Using Differentiable Knowledge Graphs</title>
@@ -5425,8 +5428,8 @@
       <url hash="43007347">2021.emnlp-main.345</url>
       <bibkey>saffari-etal-2021-end</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.345</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/simplequestions">SimpleQuestions</pwcdataset>
       <video href="2021.emnlp-main.345.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/simplequestions">SimpleQuestions</pwcdataset>
     </paper>
     <paper id="346">
       <title>Improving Query Graph Generation for Complex Question Answering over Knowledge Base</title>
@@ -5439,8 +5442,8 @@
       <url hash="54ebb3cd">2021.emnlp-main.346</url>
       <bibkey>qin-etal-2021-improving-query</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.346</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/simplequestions">SimpleQuestions</pwcdataset>
       <video href="2021.emnlp-main.346.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/simplequestions">SimpleQuestions</pwcdataset>
     </paper>
     <paper id="347">
       <title><fixed-case>D</fixed-case>isco<fixed-case>DVT</fixed-case>: <fixed-case>G</fixed-case>enerating Long Text with Discourse-Aware Discrete Variational Transformer</title>
@@ -5451,10 +5454,10 @@
       <url hash="758ae64b">2021.emnlp-main.347</url>
       <bibkey>ji-huang-2021-discodvt</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.347</doi>
+      <video href="2021.emnlp-main.347.mp4"/>
       <pwccode url="https://github.com/cdjhz/discodvt" additional="false">cdjhz/discodvt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
-      <video href="2021.emnlp-main.347.mp4"/>
     </paper>
     <paper id="348">
       <title>Mathematical Word Problem Generation from Commonsense Knowledge Graph and Equations</title>
@@ -5469,9 +5472,9 @@
       <url hash="170b9494">2021.emnlp-main.348</url>
       <bibkey>liu-etal-2021-mathematical</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.348</doi>
+      <video href="2021.emnlp-main.348.mp4"/>
       <pwccode url="https://github.com/tal-ai/make_emnlp2021" additional="false">tal-ai/make_emnlp2021</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mathqa">MathQA</pwcdataset>
-      <video href="2021.emnlp-main.348.mp4"/>
     </paper>
     <paper id="349">
       <title>Generic resources are what you need: Style transfer tasks without task-specific parallel training data</title>
@@ -5483,9 +5486,9 @@
       <url hash="4b8c3c7b">2021.emnlp-main.349</url>
       <bibkey>lai-etal-2021-generic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.349</doi>
+      <video href="2021.emnlp-main.349.mp4"/>
       <pwccode url="https://github.com/laihuiyuan/generic-resources-for-tst" additional="false">laihuiyuan/generic-resources-for-tst</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
-      <video href="2021.emnlp-main.349.mp4"/>
     </paper>
     <paper id="350">
       <title>Revisiting Pivot-Based Paraphrase Generation: Language Is Not the Only Optional Pivot</title>
@@ -5509,10 +5512,10 @@
       <url hash="0888b04a">2021.emnlp-main.351</url>
       <bibkey>ribeiro-etal-2021-structural</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.351</doi>
+      <video href="2021.emnlp-main.351.mp4"/>
       <pwccode url="https://github.com/ukplab/structadapt" additional="false">ukplab/structadapt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/amr3-0">AMR3.0</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ldc2020t02">LDC2020T02</pwcdataset>
-      <video href="2021.emnlp-main.351.mp4"/>
     </paper>
     <paper id="352">
       <title>Data-to-text Generation by Splicing Together Nearest Neighbors</title>
@@ -5524,9 +5527,9 @@
       <url hash="5aa404e4">2021.emnlp-main.352</url>
       <bibkey>wiseman-etal-2021-data</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.352</doi>
+      <video href="2021.emnlp-main.352.mp4"/>
       <pwccode url="https://github.com/swiseman/neighbor-splicing" additional="false">swiseman/neighbor-splicing</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikibio">WikiBio</pwcdataset>
-      <video href="2021.emnlp-main.352.mp4"/>
     </paper>
     <paper id="353">
       <title>Contextualize Knowledge Bases with Transformer for End-to-end Task-Oriented Dialogue Systems</title>
@@ -5565,9 +5568,9 @@
       <url hash="df1167d3">2021.emnlp-main.355</url>
       <bibkey>zhou-etal-2021-crfr</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.355</doi>
+      <video href="2021.emnlp-main.355.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/redial">ReDial</pwcdataset>
-      <video href="2021.emnlp-main.355.mp4"/>
     </paper>
     <paper id="356">
       <title><fixed-case>D</fixed-case>u<fixed-case>R</fixed-case>ec<fixed-case>D</fixed-case>ial 2.0: A Bilingual Parallel Corpus for Conversational Recommendation</title>
@@ -5581,9 +5584,9 @@
       <url hash="d8bb4476">2021.emnlp-main.356</url>
       <bibkey>liu-etal-2021-durecdial</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.356</doi>
+      <video href="2021.emnlp-main.356.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/durecdial">DuRecDial</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/redial">ReDial</pwcdataset>
-      <video href="2021.emnlp-main.356.mp4"/>
     </paper>
     <paper id="357">
       <title>End-to-End Learning of Flowchart Grounded Task-Oriented Dialogs</title>
@@ -5596,9 +5599,9 @@
       <url hash="9857fbd2">2021.emnlp-main.357</url>
       <bibkey>raghu-etal-2021-end</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.357</doi>
+      <video href="2021.emnlp-main.357.mp4"/>
       <pwccode url="https://github.com/dair-iitd/flonet" additional="false">dair-iitd/flonet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/flodial">FloDial</pwcdataset>
-      <video href="2021.emnlp-main.357.mp4"/>
     </paper>
     <paper id="358">
       <title>Dimensional Emotion Detection from Categorical Emotion</title>
@@ -5613,10 +5616,10 @@
       <url hash="b0fec8db">2021.emnlp-main.358</url>
       <bibkey>park-etal-2021-dimensional</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.358</doi>
+      <video href="2021.emnlp-main.358.mp4"/>
       <pwccode url="https://github.com/sungjoonpark/emotiondetection" additional="false">sungjoonpark/emotiondetection</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/emobank">EmoBank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/isear">ISEAR</pwcdataset>
-      <video href="2021.emnlp-main.358.mp4"/>
     </paper>
     <paper id="359">
       <title>Not All Negatives are Equal: <fixed-case>L</fixed-case>abel-Aware Contrastive Loss for Fine-grained Text Classification</title>
@@ -5627,11 +5630,11 @@
       <url hash="dbdabe96">2021.emnlp-main.359</url>
       <bibkey>suresh-ong-2021-negatives</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.359</doi>
+      <video href="2021.emnlp-main.359.mp4"/>
       <pwccode url="https://github.com/varsha33/lcl_loss" additional="false">varsha33/lcl_loss</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/goemotions">GoEmotions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/isear">ISEAR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.359.mp4"/>
     </paper>
     <paper id="360">
       <title>Joint Multi-modal Aspect-Sentiment Analysis with Auxiliary Cross-modal Relation Detection</title>
@@ -5647,8 +5650,8 @@
       <url hash="84931f70">2021.emnlp-main.360</url>
       <bibkey>ju-etal-2021-joint</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.360</doi>
-      <pwccode url="https://github.com/manlp-suda/jml" additional="false">manlp-suda/jml</pwccode>
       <video href="2021.emnlp-main.360.mp4"/>
+      <pwccode url="https://github.com/manlp-suda/jml" additional="false">manlp-suda/jml</pwccode>
     </paper>
     <paper id="361">
       <title>Solving Aspect Category Sentiment Analysis as a Text Generation Task</title>
@@ -5662,9 +5665,9 @@
       <url hash="8f2fe667">2021.emnlp-main.361</url>
       <bibkey>liu-etal-2021-solving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.361</doi>
+      <video href="2021.emnlp-main.361.mp4"/>
       <pwccode url="https://github.com/lgw863/acsa-generation" additional="false">lgw863/acsa-generation</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mams">MAMS</pwcdataset>
-      <video href="2021.emnlp-main.361.mp4"/>
     </paper>
     <paper id="362">
       <title>Semantics-Preserved Data Augmentation for Aspect-Based Sentiment Analysis</title>
@@ -5678,10 +5681,10 @@
       <attachment type="Software" hash="67ccb15b">2021.emnlp-main.362.Software.zip</attachment>
       <bibkey>hsu-etal-2021-semantics</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.362</doi>
+      <video href="2021.emnlp-main.362.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/mams">MAMS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/stocknet-1">StockNet</pwcdataset>
-      <video href="2021.emnlp-main.362.mp4"/>
     </paper>
     <paper id="363">
       <title>The Effect of Round-Trip Translation on Fairness in Sentiment Analysis</title>
@@ -5721,8 +5724,8 @@
       <url hash="b0fd9833">2021.emnlp-main.365</url>
       <bibkey>lin-etal-2021-csds</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.365</doi>
-      <pwccode url="https://github.com/xiaolinAndy/CSDS" additional="false">xiaolinAndy/CSDS</pwccode>
       <video href="2021.emnlp-main.365.mp4"/>
+      <pwccode url="https://github.com/xiaolinAndy/CSDS" additional="false">xiaolinAndy/CSDS</pwccode>
     </paper>
     <paper id="366">
       <title><fixed-case>C</fixed-case>od<fixed-case>RED</fixed-case>: A Cross-Document Relation Extraction Dataset for Acquiring Knowledge in the Wild</title>
@@ -5738,12 +5741,12 @@
       <url hash="e524d1d5">2021.emnlp-main.366</url>
       <bibkey>yao-etal-2021-codred</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.366</doi>
+      <video href="2021.emnlp-main.366.mp4"/>
       <pwccode url="https://github.com/thunlp/codred" additional="false">thunlp/codred</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bc5cdr">BC5CDR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/docred">DocRED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/knowledgenet">KnowledgeNet</pwcdataset>
-      <video href="2021.emnlp-main.366.mp4"/>
     </paper>
     <paper id="367">
       <title>Building and Evaluating Open-Domain Dialogue Corpora with Clarifying Questions</title>
@@ -5757,10 +5760,10 @@
       <url hash="6ff0caec">2021.emnlp-main.367</url>
       <bibkey>aliannejadi-etal-2021-building</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.367</doi>
+      <video href="2021.emnlp-main.367.mp4"/>
       <pwccode url="https://github.com/aliannejadi/ClariQ" additional="false">aliannejadi/ClariQ</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clariq">ClariQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qulac">Qulac</pwcdataset>
-      <video href="2021.emnlp-main.367.mp4"/>
     </paper>
     <paper id="368">
       <title>We Need to Talk About train-dev-test Splits</title>
@@ -5771,8 +5774,8 @@
       <attachment type="Software" hash="8ea0c706">2021.emnlp-main.368.Software.tgz</attachment>
       <bibkey>van-der-goot-2021-need</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.368</doi>
-      <pwccode url="https://bitbucket.org/robvanderg/tuneset" additional="false">robvanderg/tuneset</pwccode>
       <video href="2021.emnlp-main.368.mp4"/>
+      <pwccode url="https://bitbucket.org/robvanderg/tuneset" additional="false">robvanderg/tuneset</pwccode>
     </paper>
     <paper id="369">
       <title><fixed-case>P</fixed-case>ho<fixed-case>MT</fixed-case>: A High-Quality and Large-Scale Benchmark Dataset for <fixed-case>V</fixed-case>ietnamese-<fixed-case>E</fixed-case>nglish Machine Translation</title>
@@ -5786,11 +5789,11 @@
       <url hash="587dc8f9">2021.emnlp-main.369</url>
       <bibkey>doan-etal-2021-phomt</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.369</doi>
+      <video href="2021.emnlp-main.369.mp4"/>
       <pwccode url="https://github.com/vinairesearch/phomt" additional="false">vinairesearch/phomt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/phomt">PhoMT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikimatrix">WikiMatrix</pwcdataset>
-      <video href="2021.emnlp-main.369.mp4"/>
     </paper>
     <paper id="370">
       <title>Lying Through One’s Teeth: A Study on Verbal Leakage Cues</title>
@@ -5813,9 +5816,9 @@
       <url hash="d50154be">2021.emnlp-main.371</url>
       <bibkey>chen-etal-2021-multi</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.371</doi>
+      <video href="2021.emnlp-main.371.mp4"/>
       <pwccode url="https://github.com/yangyi-chen/maya" additional="false">yangyi-chen/maya</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.371.mp4"/>
     </paper>
     <paper id="372">
       <title>All Bark and No Bite: Rogue Dimensions in Transformer Language Models Obscure Representational Quality</title>
@@ -5826,8 +5829,8 @@
       <url hash="a3c7ad7b">2021.emnlp-main.372</url>
       <bibkey>timkey-van-schijndel-2021-bark</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.372</doi>
-      <pwccode url="https://github.com/wtimkey/rogue-dimensions" additional="false">wtimkey/rogue-dimensions</pwccode>
       <video href="2021.emnlp-main.372.mp4"/>
+      <pwccode url="https://github.com/wtimkey/rogue-dimensions" additional="false">wtimkey/rogue-dimensions</pwccode>
     </paper>
     <paper id="373">
       <title><fixed-case>I</fixed-case>ncorporating <fixed-case>R</fixed-case>esidual and <fixed-case>N</fixed-case>ormalization <fixed-case>L</fixed-case>ayers into <fixed-case>A</fixed-case>nalysis of <fixed-case>M</fixed-case>asked <fixed-case>L</fixed-case>anguage <fixed-case>M</fixed-case>odels</title>
@@ -5840,10 +5843,10 @@
       <url hash="5af48518">2021.emnlp-main.373</url>
       <bibkey>kobayashi-etal-2021-incorporating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.373</doi>
+      <video href="2021.emnlp-main.373.mp4"/>
       <pwccode url="https://github.com/gorokoba560/norm-analysis-of-transformer" additional="true">gorokoba560/norm-analysis-of-transformer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.373.mp4"/>
     </paper>
     <paper id="374">
       <title>Mind the Style of Text! Adversarial and Backdoor Attacks Based on Text Style Transfer</title>
@@ -5858,10 +5861,10 @@
       <url hash="dd7a40e6">2021.emnlp-main.374</url>
       <bibkey>qi-etal-2021-mind</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.374</doi>
+      <video href="2021.emnlp-main.374.mp4"/>
       <pwccode url="https://github.com/thunlp/styleattack" additional="false">thunlp/styleattack</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hate-speech">Hate Speech</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.374.mp4"/>
     </paper>
     <paper id="375">
       <title>Sociolectal Analysis of Pretrained Language Models</title>
@@ -5925,8 +5928,8 @@
       <url hash="9c0a36cf">2021.emnlp-main.379</url>
       <bibkey>yang-etal-2021-enhanced</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.379</doi>
-      <pwccode url="https://github.com/akeepers/lear" additional="false">akeepers/lear</pwccode>
       <video href="2021.emnlp-main.379.mp4"/>
+      <pwccode url="https://github.com/akeepers/lear" additional="false">akeepers/lear</pwccode>
     </paper>
     <paper id="380">
       <title><fixed-case>PRIDE</fixed-case>: <fixed-case>P</fixed-case>redicting <fixed-case>R</fixed-case>elationships in <fixed-case>C</fixed-case>onversations</title>
@@ -5939,8 +5942,8 @@
       <url hash="b11b038d">2021.emnlp-main.380</url>
       <bibkey>tigunova-etal-2021-pride</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.380</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/ddrel">DDRel</pwcdataset>
       <video href="2021.emnlp-main.380.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ddrel">DDRel</pwcdataset>
     </paper>
     <paper id="381">
       <title>Extracting Fine-Grained Knowledge Graphs of Scientific Claims: Dataset and Transformer-Based Results</title>
@@ -5951,11 +5954,11 @@
       <url hash="34b1bdff">2021.emnlp-main.381</url>
       <bibkey>magnusson-friedman-2021-extracting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.381</doi>
+      <video href="2021.emnlp-main.381.mp4"/>
       <pwccode url="https://github.com/siftech/sciclaim" additional="false">siftech/sciclaim</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cord-19">CORD-19</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scirex">SciREX</pwcdataset>
-      <video href="2021.emnlp-main.381.mp4"/>
     </paper>
     <paper id="382">
       <title>Sequential Cross-Document Coreference Resolution</title>
@@ -5967,8 +5970,8 @@
       <url hash="0e3d612c">2021.emnlp-main.382</url>
       <bibkey>allaway-etal-2021-sequential</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.382</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/ecb">ECB+</pwcdataset>
       <video href="2021.emnlp-main.382.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ecb">ECB+</pwcdataset>
     </paper>
     <paper id="383">
       <title>Mixture-of-Partitions: Infusing Large Biomedical Knowledge Graphs into <fixed-case>BERT</fixed-case></title>
@@ -5982,9 +5985,9 @@
       <url hash="e8e89df6">2021.emnlp-main.383</url>
       <bibkey>meng-etal-2021-mixture</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.383</doi>
+      <video href="2021.emnlp-main.383.mp4"/>
       <pwccode url="https://github.com/cambridgeltl/mop" additional="false">cambridgeltl/mop</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/pubmedqa">PubMedQA</pwcdataset>
-      <video href="2021.emnlp-main.383.mp4"/>
     </paper>
     <paper id="384">
       <title>Filling the Gaps in <fixed-case>A</fixed-case>ncient <fixed-case>A</fixed-case>kkadian Texts: A Masked Language Modelling Approach</title>
@@ -5999,8 +6002,8 @@
       <url hash="f27776de">2021.emnlp-main.384</url>
       <bibkey>lazar-etal-2021-filling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.384</doi>
-      <pwccode url="https://github.com/slab-nlp/akk" additional="false">slab-nlp/akk</pwccode>
       <video href="2021.emnlp-main.384.mp4"/>
+      <pwccode url="https://github.com/slab-nlp/akk" additional="false">slab-nlp/akk</pwccode>
     </paper>
     <paper id="385">
       <title><fixed-case>AV</fixed-case>oca<fixed-case>D</fixed-case>o: Strategy for Adapting Vocabulary to Downstream Domain</title>
@@ -6013,8 +6016,8 @@
       <url hash="d7984f5f">2021.emnlp-main.385</url>
       <bibkey>hong-etal-2021-avocado</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.385</doi>
-      <pwccode url="https://github.com/Jimin9401/avocado" additional="false">Jimin9401/avocado</pwccode>
       <video href="2021.emnlp-main.385.mp4"/>
+      <pwccode url="https://github.com/Jimin9401/avocado" additional="false">Jimin9401/avocado</pwccode>
     </paper>
     <paper id="386">
       <title>Can We Improve Model Robustness through Secondary Attribute Counterfactuals?</title>
@@ -6046,9 +6049,9 @@
       <url hash="d6f47389">2021.emnlp-main.387</url>
       <bibkey>clement-etal-2021-long</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.387</doi>
+      <video href="2021.emnlp-main.387.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/codesearchnet">CodeSearchNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/codexglue">CodeXGLUE</pwcdataset>
-      <video href="2021.emnlp-main.387.mp4"/>
     </paper>
     <paper id="388">
       <title>Can Language Models be Biomedical Knowledge Bases?</title>
@@ -6063,10 +6066,10 @@
       <url hash="497056c4">2021.emnlp-main.388</url>
       <bibkey>sung-etal-2021-language</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.388</doi>
+      <video href="2021.emnlp-main.388.mp4"/>
       <pwccode url="https://github.com/dmis-lab/biolama" additional="false">dmis-lab/biolama</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/biolama">BioLAMA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
-      <video href="2021.emnlp-main.388.mp4"/>
     </paper>
     <paper id="389">
       <title><fixed-case>L</fixed-case>ayout<fixed-case>R</fixed-case>eader: Pre-training of Text and Layout for Reading Order Detection</title>
@@ -6080,9 +6083,9 @@
       <url hash="ad71e00b">2021.emnlp-main.389</url>
       <bibkey>wang-etal-2021-layoutreader</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.389</doi>
+      <video href="2021.emnlp-main.389.mp4"/>
       <pwccode url="https://github.com/microsoft/unilm" additional="false">microsoft/unilm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/readingbank">ReadingBank</pwcdataset>
-      <video href="2021.emnlp-main.389.mp4"/>
     </paper>
     <paper id="390">
       <title>Region under <fixed-case>D</fixed-case>iscussion for visual dialog</title>
@@ -6098,10 +6101,10 @@
       <attachment type="Software" hash="a12b6638">2021.emnlp-main.390.Software.zip</attachment>
       <bibkey>mazuecos-etal-2021-region</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.390</doi>
+      <video href="2021.emnlp-main.390.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/guesswhat">GuessWhat?!</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visdial">VisDial</pwcdataset>
-      <video href="2021.emnlp-main.390.mp4"/>
     </paper>
     <paper id="391">
       <title>Learning grounded word meaning representations on similarity graphs</title>
@@ -6113,8 +6116,8 @@
       <url hash="e5d05ba7">2021.emnlp-main.391</url>
       <bibkey>dimiccoli-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.391</doi>
-      <pwccode url="https://github.com/mdimiccoli/hm-sge" additional="false">mdimiccoli/hm-sge</pwccode>
       <video href="2021.emnlp-main.391.mp4"/>
+      <pwccode url="https://github.com/mdimiccoli/hm-sge" additional="false">mdimiccoli/hm-sge</pwccode>
     </paper>
     <paper id="392">
       <title><fixed-case>W</fixed-case>hy<fixed-case>A</fixed-case>ct: Identifying Action Reasons in Lifestyle Vlogs</title>
@@ -6128,13 +6131,13 @@
       <url hash="da0ec43a">2021.emnlp-main.392</url>
       <bibkey>ignat-etal-2021-whyact</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.392</doi>
+      <video href="2021.emnlp-main.392.mp4"/>
       <pwccode url="https://github.com/michigannlp/vlog_action_reason" additional="false">michigannlp/vlog_action_reason</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/whyact">WhyAct</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vcr">VCR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vlogs">Vlogs</pwcdataset>
-      <video href="2021.emnlp-main.392.mp4"/>
     </paper>
     <paper id="393">
       <title>Genre as Weak Supervision for Cross-lingual Dependency Parsing</title>
@@ -6146,9 +6149,9 @@
       <url hash="71deb6c1">2021.emnlp-main.393</url>
       <bibkey>muller-eberstein-etal-2021-genre</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.393</doi>
+      <video href="2021.emnlp-main.393.mp4"/>
       <pwccode url="https://github.com/personads/ud-selection" additional="false">personads/ud-selection</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.emnlp-main.393.mp4"/>
     </paper>
     <paper id="394">
       <title>On the Relation between Syntactic Divergence and Zero-Shot Performance</title>
@@ -6161,10 +6164,10 @@
       <url hash="29f721bf">2021.emnlp-main.394</url>
       <bibkey>arviv-etal-2021-relation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.394</doi>
+      <video href="2021.emnlp-main.394.mp4"/>
       <pwccode url="https://github.com/ofirarviv/improving-ud" additional="false">ofirarviv/improving-ud</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/translated-tacred">Translated TACRED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.emnlp-main.394.mp4"/>
     </paper>
     <paper id="395">
       <title>Improved Latent Tree Induction with Distant Supervision via Span Constraints</title>
@@ -6182,10 +6185,10 @@
       <url hash="02716d62">2021.emnlp-main.395</url>
       <bibkey>xu-etal-2021-improved</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.395</doi>
+      <video href="2021.emnlp-main.395.mp4"/>
       <pwccode url="https://github.com/iesl/distantly-supervised-diora" additional="false">iesl/distantly-supervised-diora</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/medmentions">MedMentions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.emnlp-main.395.mp4"/>
     </paper>
     <paper id="396">
       <title>Aligning Multidimensional Worldviews and Discovering Ideological Differences</title>
@@ -6210,9 +6213,9 @@
       <url hash="f240bf88">2021.emnlp-main.397</url>
       <bibkey>baheti-etal-2021-just</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.397</doi>
+      <video href="2021.emnlp-main.397.mp4"/>
       <pwccode url="https://github.com/abaheti95/toxichat" additional="false">abaheti95/toxichat</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sbic">SBIC</pwcdataset>
-      <video href="2021.emnlp-main.397.mp4"/>
     </paper>
     <paper id="398">
       <title>Multi-Modal Open-Domain Dialogue</title>
@@ -6225,12 +6228,12 @@
       <url hash="9bfa83f0">2021.emnlp-main.398</url>
       <bibkey>shuster-etal-2021-multi</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.398</doi>
+      <video href="2021.emnlp-main.398.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/blended-skill-talk">Blended Skill Talk</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coco-captions">COCO Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/convai2">ConvAI2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/empatheticdialogues">EmpatheticDialogues</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.emnlp-main.398.mp4"/>
     </paper>
     <paper id="399">
       <title>A Label-Aware <fixed-case>BERT</fixed-case> Attention Network for Zero-Shot Multi-Intent Detection in Spoken Language Understanding</title>
@@ -6242,10 +6245,10 @@
       <url hash="07f4a1fb">2021.emnlp-main.399</url>
       <bibkey>wu-etal-2021-label</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.399</doi>
+      <video href="2021.emnlp-main.399.mp4"/>
       <pwccode url="https://github.com/waynewu6250/laban" additional="false">waynewu6250/laban</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sgd">SGD</pwcdataset>
-      <video href="2021.emnlp-main.399.mp4"/>
     </paper>
     <paper id="400">
       <title>Zero-Shot Dialogue Disentanglement by Self-Supervised Entangled Response Selection</title>
@@ -6256,8 +6259,8 @@
       <url hash="3cd15251">2021.emnlp-main.400</url>
       <bibkey>chi-rudnicky-2021-zero</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.400</doi>
-      <pwccode url="https://github.com/chijames/zero_shot_dialogue_disentanglement" additional="false">chijames/zero_shot_dialogue_disentanglement</pwccode>
       <video href="2021.emnlp-main.400.mp4"/>
+      <pwccode url="https://github.com/chijames/zero_shot_dialogue_disentanglement" additional="false">chijames/zero_shot_dialogue_disentanglement</pwccode>
     </paper>
     <paper id="401">
       <title><fixed-case>SIMMC</fixed-case> 2.0: A Task-oriented Dialog Dataset for Immersive Multimodal Conversations</title>
@@ -6271,10 +6274,10 @@
       <attachment type="Software" hash="f0f9c662">2021.emnlp-main.401.Software.zip</attachment>
       <bibkey>kottur-etal-2021-simmc</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.401</doi>
+      <video href="2021.emnlp-main.401.mp4"/>
       <pwccode url="https://github.com/facebookresearch/simmc2" additional="false">facebookresearch/simmc2</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/simmc">SIMMC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.emnlp-main.401.mp4"/>
     </paper>
     <paper id="402">
       <title><fixed-case>RAST</fixed-case>: Domain-Robust Dialogue Rewriting as Sequence Tagging</title>
@@ -6302,10 +6305,10 @@
       <url hash="b120ce70">2021.emnlp-main.403</url>
       <bibkey>grover-etal-2021-mrf</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.403</doi>
+      <video href="2021.emnlp-main.403.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/blended-skill-talk">Blended Skill Talk</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/convai2">ConvAI2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.emnlp-main.403.mp4"/>
     </paper>
     <paper id="404">
       <title>Dialogue State Tracking with a Language Model using Schema-Driven Prompting</title>
@@ -6317,8 +6320,8 @@
       <url hash="8c222adc">2021.emnlp-main.404</url>
       <bibkey>lee-etal-2021-dialogue</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.404</doi>
-      <pwccode url="https://github.com/chiahsuan156/dst-as-prompting" additional="false">chiahsuan156/dst-as-prompting</pwccode>
       <video href="2021.emnlp-main.404.mp4"/>
+      <pwccode url="https://github.com/chiahsuan156/dst-as-prompting" additional="false">chiahsuan156/dst-as-prompting</pwccode>
     </paper>
     <paper id="405">
       <title>Signed Coreference Resolution</title>
@@ -6330,8 +6333,8 @@
       <url hash="d911a424">2021.emnlp-main.405</url>
       <bibkey>yin-etal-2021-signed</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.405</doi>
-      <pwccode url="https://github.com/kayoyin/scr" additional="false">kayoyin/scr</pwccode>
       <video href="2021.emnlp-main.405.mp4"/>
+      <pwccode url="https://github.com/kayoyin/scr" additional="false">kayoyin/scr</pwccode>
     </paper>
     <paper id="406">
       <title>Consistent Accelerated Inference via Confident Adaptive Transformers</title>
@@ -6344,10 +6347,10 @@
       <url hash="cd2ad2bf">2021.emnlp-main.406</url>
       <bibkey>schuster-etal-2021-consistent</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.406</doi>
+      <video href="2021.emnlp-main.406.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vitaminc">VitaminC</pwcdataset>
-      <video href="2021.emnlp-main.406.mp4"/>
     </paper>
     <paper id="407">
       <title>Improving and Simplifying Pattern Exploiting Training</title>
@@ -6362,6 +6365,7 @@
       <attachment type="Software" hash="329004aa">2021.emnlp-main.407.Software.zip</attachment>
       <bibkey>tam-etal-2021-improving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.407</doi>
+      <video href="2021.emnlp-main.407.mp4"/>
       <pwccode url="https://github.com/rrmenon10/ADAPET" additional="true">rrmenon10/ADAPET</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
@@ -6370,7 +6374,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
-      <video href="2021.emnlp-main.407.mp4"/>
     </paper>
     <paper id="408">
       <title>Unsupervised Data Augmentation with Naive Augmentation and without Unlabeled Data</title>
@@ -6383,10 +6386,10 @@
       <url hash="ec528452">2021.emnlp-main.408</url>
       <bibkey>lowell-etal-2021-unsupervised</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.408</doi>
+      <video href="2021.emnlp-main.408.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/evidence-inference">Evidence Inference</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.408.mp4"/>
     </paper>
     <paper id="409">
       <title>Pre-train or Annotate? Domain Adaptation with a Constrained Budget</title>
@@ -6398,8 +6401,8 @@
       <url hash="7276f144">2021.emnlp-main.409</url>
       <bibkey>bai-etal-2021-pre</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.409</doi>
-      <pwccode url="https://github.com/bflashcp3f/procbert" additional="false">bflashcp3f/procbert</pwccode>
       <video href="2021.emnlp-main.409.mp4"/>
+      <pwccode url="https://github.com/bflashcp3f/procbert" additional="false">bflashcp3f/procbert</pwccode>
     </paper>
     <paper id="410">
       <title>Lawyers are Dishonest? Quantifying Representational Harms in Commonsense Knowledge Resources</title>
@@ -6414,9 +6417,9 @@
       <url hash="a97613c9">2021.emnlp-main.410</url>
       <bibkey>mehrabi-etal-2021-lawyers</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.410</doi>
+      <video href="2021.emnlp-main.410.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/genericskb">GenericsKB</pwcdataset>
-      <video href="2021.emnlp-main.410.mp4"/>
     </paper>
     <paper id="411">
       <title><fixed-case>OSC</fixed-case>a<fixed-case>R</fixed-case>: Orthogonal Subspace Correction and Rectification of Biases in Word Embeddings</title>
@@ -6429,10 +6432,10 @@
       <url hash="5bd5e3a3">2021.emnlp-main.411</url>
       <bibkey>dev-etal-2021-oscar</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.411</doi>
+      <video href="2021.emnlp-main.411.mp4"/>
       <pwccode url="https://github.com/Shaul1321/nullspace_projection" additional="false">Shaul1321/nullspace_projection</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.411.mp4"/>
     </paper>
     <paper id="412">
       <title>Sentence-Permuted Paragraph Generation</title>
@@ -6446,10 +6449,10 @@
       <url hash="0346f9b4">2021.emnlp-main.412</url>
       <bibkey>yu-etal-2021-sentence</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.412</doi>
+      <video href="2021.emnlp-main.412.mp4"/>
       <pwccode url="https://github.com/wyu97/permgen" additional="false">wyu97/permgen</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/agenda">AGENDA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
-      <video href="2021.emnlp-main.412.mp4"/>
     </paper>
     <paper id="413">
       <title>Extract, Denoise and Enforce: Evaluating and Improving Concept Preservation for Text-to-Text Generation</title>
@@ -6463,9 +6466,9 @@
       <url hash="d6884d47">2021.emnlp-main.413</url>
       <bibkey>mao-etal-2021-extract</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.413</doi>
+      <video href="2021.emnlp-main.413.mp4"/>
       <pwccode url="https://github.com/morningmoni/ede" additional="true">morningmoni/ede</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.413.mp4"/>
     </paper>
     <paper id="414">
       <title>Paraphrase Generation: A Survey of the State of the Art</title>
@@ -6476,8 +6479,8 @@
       <url hash="fc8044b8">2021.emnlp-main.414</url>
       <bibkey>zhou-bhat-2021-paraphrase</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.414</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <video href="2021.emnlp-main.414.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
     </paper>
     <paper id="415">
       <title>Exposure Bias versus Self-Recovery: Are Distortions Really Incremental for Autoregressive Text Generation?</title>
@@ -6503,11 +6506,11 @@
       <url hash="69962c6b">2021.emnlp-main.416</url>
       <bibkey>zhou-etal-2021-generating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.416</doi>
+      <video href="2021.emnlp-main.416.mp4"/>
       <pwccode url="https://github.com/amazon-research/sc2qa-dril" additional="false">amazon-research/sc2qa-dril</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsqa">NewsQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.416.mp4"/>
     </paper>
     <paper id="417">
       <title>Unsupervised Paraphrasing with Pretrained Language Models</title>
@@ -6549,8 +6552,8 @@
       <url hash="75b8ffee">2021.emnlp-main.419</url>
       <bibkey>yang-etal-2021-journalistic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.419</doi>
-      <pwccode url="https://github.com/dataminr-ai/joganic" additional="false">dataminr-ai/joganic</pwccode>
       <video href="2021.emnlp-main.419.mp4"/>
+      <pwccode url="https://github.com/dataminr-ai/joganic" additional="false">dataminr-ai/joganic</pwccode>
     </paper>
     <paper id="420">
       <title><fixed-case>AESOP</fixed-case>: Paraphrase Generation with Adaptive Syntactic Control</title>
@@ -6562,9 +6565,9 @@
       <url hash="9026394d">2021.emnlp-main.420</url>
       <bibkey>sun-etal-2021-aesop</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.420</doi>
+      <video href="2021.emnlp-main.420.mp4"/>
       <pwccode url="https://github.com/pluslabnlp/aesop" additional="false">pluslabnlp/aesop</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
-      <video href="2021.emnlp-main.420.mp4"/>
     </paper>
     <paper id="421">
       <title>Refocusing on Relevance: Personalization in <fixed-case>NLG</fixed-case></title>
@@ -6576,8 +6579,8 @@
       <url hash="6ba7f724">2021.emnlp-main.421</url>
       <bibkey>dudy-etal-2021-refocusing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.421</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/mimic-iii">MIMIC-III</pwcdataset>
       <video href="2021.emnlp-main.421.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mimic-iii">MIMIC-III</pwcdataset>
     </paper>
     <paper id="422">
       <title>The Future is not One-dimensional: Complex Event Schema Induction by Graph Modeling for Event Prediction</title>
@@ -6594,8 +6597,8 @@
       <url hash="c55fb007">2021.emnlp-main.422</url>
       <bibkey>li-etal-2021-future</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.422</doi>
-      <pwccode url="https://github.com/limanling/temporal-graph-schema" additional="false">limanling/temporal-graph-schema</pwccode>
       <video href="2021.emnlp-main.422.mp4"/>
+      <pwccode url="https://github.com/limanling/temporal-graph-schema" additional="false">limanling/temporal-graph-schema</pwccode>
     </paper>
     <paper id="423">
       <title>Learning Constraints and Descriptive Segmentation for Subevent Detection</title>
@@ -6633,9 +6636,9 @@
       <url hash="c81b861b">2021.emnlp-main.425</url>
       <bibkey>xia-van-durme-2021-moving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.425</doi>
+      <video href="2021.emnlp-main.425.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/gap-coreference-dataset">GAP Coreference Dataset</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/preco">PreCo</pwcdataset>
-      <video href="2021.emnlp-main.425.mp4"/>
     </paper>
     <paper id="426">
       <title>Document-level Entity-based Extraction as Template Generation</title>
@@ -6647,10 +6650,10 @@
       <url hash="ed2611c2">2021.emnlp-main.426</url>
       <bibkey>huang-etal-2021-document</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.426</doi>
+      <video href="2021.emnlp-main.426.mp4"/>
       <pwccode url="https://github.com/PlusLabNLP/TempGen" additional="false">PlusLabNLP/TempGen</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/muc-4">MUC-4</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scirex">SciREX</pwcdataset>
-      <video href="2021.emnlp-main.426.mp4"/>
     </paper>
     <paper id="427">
       <title>Learning Prototype Representations Across Few-Shot Tasks for Event Detection</title>
@@ -6662,8 +6665,8 @@
       <url hash="f5bce283">2021.emnlp-main.427</url>
       <bibkey>lai-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.427</doi>
-      <pwccode url="https://github.com/laiviet/fsl-proact" additional="false">laiviet/fsl-proact</pwccode>
       <video href="2021.emnlp-main.427.mp4"/>
+      <pwccode url="https://github.com/laiviet/fsl-proact" additional="false">laiviet/fsl-proact</pwccode>
     </paper>
     <paper id="428">
       <title>Lifelong Event Detection with Knowledge Transfer</title>
@@ -6675,9 +6678,9 @@
       <url hash="a6ec3397">2021.emnlp-main.428</url>
       <bibkey>yu-etal-2021-lifelong</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.428</doi>
+      <video href="2021.emnlp-main.428.mp4"/>
       <pwccode url="https://github.com/perfec-yu/lifelong-ed" additional="false">perfec-yu/lifelong-ed</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/maven">MAVEN</pwcdataset>
-      <video href="2021.emnlp-main.428.mp4"/>
     </paper>
     <paper id="429">
       <title>Modular Self-Supervision for Document-Level Relation Extraction</title>
@@ -6692,8 +6695,8 @@
       <url hash="5c979ffa">2021.emnlp-main.429</url>
       <bibkey>zhang-etal-2021-modular</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.429</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/docred">DocRED</pwcdataset>
       <video href="2021.emnlp-main.429.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/docred">DocRED</pwcdataset>
     </paper>
     <paper id="430">
       <title>Unsupervised Paraphrasing Consistency Training for Low Resource Named Entity Recognition</title>
@@ -6719,9 +6722,9 @@
       <url hash="4ab86250">2021.emnlp-main.431</url>
       <bibkey>qian-etal-2021-fine</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.431</doi>
+      <video href="2021.emnlp-main.431.mp4"/>
       <pwccode url="https://github.com/lemaoliu/fet-data" additional="false">lemaoliu/fet-data</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/figer">FIGER</pwcdataset>
-      <video href="2021.emnlp-main.431.mp4"/>
     </paper>
     <paper id="432">
       <title>Adversarial Attack against Cross-lingual Knowledge Graph Alignment</title>
@@ -6751,10 +6754,10 @@
       <url hash="a4479588">2021.emnlp-main.433</url>
       <bibkey>brody-etal-2021-towards</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.433</doi>
+      <video href="2021.emnlp-main.433.mp4"/>
       <pwccode url="https://github.com/bloomberg/emnlp21_fewrel" additional="false">bloomberg/emnlp21_fewrel</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel-2-0">FewRel 2.0</pwcdataset>
-      <video href="2021.emnlp-main.433.mp4"/>
     </paper>
     <paper id="434">
       <title>Data Augmentation for Cross-Domain Named Entity Recognition</title>
@@ -6767,8 +6770,8 @@
       <url hash="b067fba8">2021.emnlp-main.434</url>
       <bibkey>chen-etal-2021-data</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.434</doi>
-      <pwccode url="https://github.com/ritual-uh/style_ner" additional="false">ritual-uh/style_ner</pwccode>
       <video href="2021.emnlp-main.434.mp4"/>
+      <pwccode url="https://github.com/ritual-uh/style_ner" additional="false">ritual-uh/style_ner</pwccode>
     </paper>
     <paper id="435">
       <title>Incorporating medical knowledge in <fixed-case>BERT</fixed-case> for clinical relation extraction</title>
@@ -6779,9 +6782,9 @@
       <url hash="f7c80bb4">2021.emnlp-main.435</url>
       <bibkey>roy-pan-2021-incorporating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.435</doi>
+      <video href="2021.emnlp-main.435.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/blue">BLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/umls">UMLS</pwcdataset>
-      <video href="2021.emnlp-main.435.mp4"/>
     </paper>
     <paper id="436">
       <title><fixed-case>ECONET</fixed-case>: Effective Continual Pretraining of Language Models for Event Temporal Reasoning</title>
@@ -6793,10 +6796,10 @@
       <url hash="a82f1c78">2021.emnlp-main.436</url>
       <bibkey>han-etal-2021-econet</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.436</doi>
+      <video href="2021.emnlp-main.436.mp4"/>
       <pwccode url="https://github.com/pluslabnlp/econet" additional="true">pluslabnlp/econet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mc-taco">MC-TACO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/torque">Torque</pwcdataset>
-      <video href="2021.emnlp-main.436.mp4"/>
     </paper>
     <paper id="437">
       <title>Learning from Noisy Labels for Entity-Centric Information Extraction</title>
@@ -6807,11 +6810,11 @@
       <url hash="0bcb5d2b">2021.emnlp-main.437</url>
       <bibkey>zhou-chen-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.437</doi>
+      <video href="2021.emnlp-main.437.mp4"/>
       <pwccode url="https://github.com/wzhouad/NLL-IE" additional="false">wzhouad/NLL-IE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll">CoNLL++</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tacred">TACRED</pwcdataset>
-      <video href="2021.emnlp-main.437.mp4"/>
     </paper>
     <paper id="438">
       <title>Extracting Material Property Measurement Data from Scientific Articles</title>
@@ -6824,8 +6827,8 @@
       <url hash="ecb1e4a0">2021.emnlp-main.438</url>
       <bibkey>panapitiya-etal-2021-extracting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.438</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/s2orc">S2ORC</pwcdataset>
       <video href="2021.emnlp-main.438.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/s2orc">S2ORC</pwcdataset>
     </paper>
     <paper id="439">
       <title>Modeling Document-Level Context for Event Detection via Important Context Selection</title>
@@ -6865,9 +6868,9 @@
       <url hash="08ef28ef">2021.emnlp-main.441</url>
       <bibkey>shen-etal-2021-corpus</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.441</doi>
+      <video href="2021.emnlp-main.441.mp4"/>
       <pwccode url="https://github.com/mickeystroller/etypeclus" additional="false">mickeystroller/etypeclus</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
-      <video href="2021.emnlp-main.441.mp4"/>
     </paper>
     <paper id="442">
       <title><fixed-case>PDALN</fixed-case>: Progressive Domain Adaptation over a Pre-trained Model for Low-Resource Cross-Domain Named Entity Recognition</title>
@@ -6881,8 +6884,8 @@
       <url hash="47507954">2021.emnlp-main.442</url>
       <bibkey>zhang-etal-2021-pdaln</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.442</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/wnut-2016-ner">WNUT 2016 NER</pwcdataset>
       <video href="2021.emnlp-main.442.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wnut-2016-ner">WNUT 2016 NER</pwcdataset>
     </paper>
     <paper id="443">
       <title>Multi-Vector Attention Models for Deep Re-ranking</title>
@@ -6893,8 +6896,8 @@
       <url hash="2ad032fa">2021.emnlp-main.443</url>
       <bibkey>zhou-devlin-2021-multi</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.443</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <video href="2021.emnlp-main.443.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
     </paper>
     <paper id="444">
       <title>Toward Deconfounding the Effect of Entity Demographics for Question Answering Accuracy</title>
@@ -6906,10 +6909,10 @@
       <url hash="fc063721">2021.emnlp-main.444</url>
       <bibkey>gor-etal-2021-toward</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.444</doi>
+      <video href="2021.emnlp-main.444.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.emnlp-main.444.mp4"/>
     </paper>
     <paper id="445">
       <title>Exploring Strategies for Generalizable Commonsense Reasoning with Pre-trained Models</title>
@@ -6924,10 +6927,10 @@
       <url hash="96ae37b4">2021.emnlp-main.445</url>
       <bibkey>ma-etal-2021-exploring</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.445</doi>
+      <video href="2021.emnlp-main.445.mp4"/>
       <pwccode url="https://github.com/mayer123/cs_model_adaptation" additional="false">mayer123/cs_model_adaptation</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/commongen">CommonGen</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.445.mp4"/>
     </paper>
     <paper id="446">
       <title>Transformer Feed-Forward Layers Are Key-Value Memories</title>
@@ -6940,9 +6943,10 @@
       <url hash="d2a3c43d">2021.emnlp-main.446</url>
       <bibkey>geva-etal-2021-transformer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.446</doi>
+      <video href="2021.emnlp-main.446.mp4"/>
       <pwccode url="https://github.com/mega002/ff-layers" additional="false">mega002/ff-layers</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.emnlp-main.446.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="447">
       <title>Connecting Attributions and <fixed-case>QA</fixed-case> Model Behavior on Realistic Counterfactuals</title>
@@ -6954,10 +6958,10 @@
       <url hash="3e8b1385">2021.emnlp-main.447</url>
       <bibkey>ye-etal-2021-connecting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.447</doi>
+      <video href="2021.emnlp-main.447.mp4"/>
       <pwccode url="https://github.com/xiye17/EvalQAExpl" additional="false">xiye17/EvalQAExpl</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.447.mp4"/>
     </paper>
     <paper id="448">
       <title>How Do Neural Sequence Models Generalize? Local and Global Cues for Out-of-Distribution Prediction</title>
@@ -6979,12 +6983,12 @@
       <url hash="6fe4fed0">2021.emnlp-main.449</url>
       <bibkey>yauney-mimno-2021-comparing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.449</doi>
+      <video href="2021.emnlp-main.449.mp4"/>
       <pwccode url="https://github.com/gyauney/data-label-alignment" additional="false">gyauney/data-label-alignment</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mnist">MNIST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.449.mp4"/>
     </paper>
     <paper id="450">
       <title>Human Rationales as Attribution Priors for Explainable Stance Detection</title>
@@ -6995,8 +6999,8 @@
       <url hash="dfcee043">2021.emnlp-main.450</url>
       <bibkey>jayaram-allaway-2021-human</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.450</doi>
-      <pwccode url="https://github.com/sahilj97/explainable-stance-detection" additional="false">sahilj97/explainable-stance-detection</pwccode>
       <video href="2021.emnlp-main.450.mp4"/>
+      <pwccode url="https://github.com/sahilj97/explainable-stance-detection" additional="false">sahilj97/explainable-stance-detection</pwccode>
     </paper>
     <paper id="451">
       <title>The Stem Cell Hypothesis: Dilemma behind Multi-Task Learning with Transformer Encoders</title>
@@ -7007,8 +7011,8 @@
       <url hash="d74d25cc">2021.emnlp-main.451</url>
       <bibkey>he-choi-2021-stem</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.451</doi>
-      <pwccode url="https://github.com/emorynlp/stem-cell-hypothesis" additional="false">emorynlp/stem-cell-hypothesis</pwccode>
       <video href="2021.emnlp-main.451.mp4"/>
+      <pwccode url="https://github.com/emorynlp/stem-cell-hypothesis" additional="false">emorynlp/stem-cell-hypothesis</pwccode>
     </paper>
     <paper id="452">
       <title>Text Counterfactuals via Latent Optimization and <fixed-case>S</fixed-case>hapley-Guided Search</title>
@@ -7020,9 +7024,9 @@
       <attachment type="Software" hash="afb5d814">2021.emnlp-main.452.Software.zip</attachment>
       <bibkey>fern-pope-2021-text</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.452</doi>
+      <video href="2021.emnlp-main.452.mp4"/>
       <pwccode url="https://github.com/QuintinPope/CLOSS" additional="false">QuintinPope/CLOSS</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.452.mp4"/>
     </paper>
     <paper id="453">
       <title>“Average” Approximates “First Principal Component”? An Empirical Analysis on Representations from Neural Language Models</title>
@@ -7034,8 +7038,8 @@
       <url hash="ffa0cd89">2021.emnlp-main.453</url>
       <bibkey>wang-etal-2021-average</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.453</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
       <video href="2021.emnlp-main.453.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
     </paper>
     <paper id="454">
       <title>Controlled Evaluation of Grammatical Knowledge in <fixed-case>M</fixed-case>andarin <fixed-case>C</fixed-case>hinese Language Models</title>
@@ -7048,8 +7052,8 @@
       <url hash="f059d76d">2021.emnlp-main.454</url>
       <bibkey>wang-etal-2021-controlled</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.454</doi>
-      <pwccode url="https://github.com/yiwenwang03/syntactic-generalization-mandarin" additional="false">yiwenwang03/syntactic-generalization-mandarin</pwccode>
       <video href="2021.emnlp-main.454.mp4"/>
+      <pwccode url="https://github.com/yiwenwang03/syntactic-generalization-mandarin" additional="false">yiwenwang03/syntactic-generalization-mandarin</pwccode>
     </paper>
     <paper id="455">
       <title><fixed-case>G</fixed-case>rad<fixed-case>TS</fixed-case>: A Gradient-Based Automatic Auxiliary Task Selection Method Based on Transformer Networks</title>
@@ -7063,10 +7067,10 @@
       <url hash="42b49de4">2021.emnlp-main.455</url>
       <bibkey>ma-etal-2021-gradts</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.455</doi>
+      <video href="2021.emnlp-main.455.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.emnlp-main.455.mp4"/>
     </paper>
     <paper id="456">
       <title><fixed-case>N</fixed-case>egat<fixed-case>ER</fixed-case>: <fixed-case>U</fixed-case>nsupervised <fixed-case>D</fixed-case>iscovery of <fixed-case>N</fixed-case>egatives in <fixed-case>C</fixed-case>ommonsense <fixed-case>K</fixed-case>nowledge <fixed-case>B</fixed-case>ases</title>
@@ -7078,8 +7082,8 @@
       <url hash="0d09cf6c">2021.emnlp-main.456</url>
       <bibkey>safavi-etal-2021-negater</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.456</doi>
-      <pwccode url="https://github.com/tsafavi/negater" additional="false">tsafavi/negater</pwccode>
       <video href="2021.emnlp-main.456.mp4"/>
+      <pwccode url="https://github.com/tsafavi/negater" additional="false">tsafavi/negater</pwccode>
     </paper>
     <paper id="457">
       <title>Instance-adaptive training with noise-robust losses against noisy labels</title>
@@ -7092,9 +7096,9 @@
       <url hash="4d50790d">2021.emnlp-main.457</url>
       <bibkey>jin-etal-2021-instance</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.457</doi>
+      <video href="2021.emnlp-main.457.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.457.mp4"/>
     </paper>
     <paper id="458">
       <title>Distributionally Robust Multilingual Machine Translation</title>
@@ -7108,8 +7112,8 @@
       <url hash="86b0c8df">2021.emnlp-main.458</url>
       <bibkey>zhou-etal-2021-distributionally</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.458</doi>
-      <pwccode url="https://github.com/violet-zct/fairseq-dro-mnmt" additional="false">violet-zct/fairseq-dro-mnmt</pwccode>
       <video href="2021.emnlp-main.458.mp4"/>
+      <pwccode url="https://github.com/violet-zct/fairseq-dro-mnmt" additional="false">violet-zct/fairseq-dro-mnmt</pwccode>
     </paper>
     <paper id="459">
       <title>Model Selection for Cross-lingual Transfer</title>
@@ -7120,8 +7124,8 @@
       <url hash="48886322">2021.emnlp-main.459</url>
       <bibkey>chen-ritter-2021-model</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.459</doi>
-      <pwccode url="https://github.com/edchengg/model_selection" additional="false">edchengg/model_selection</pwccode>
       <video href="2021.emnlp-main.459.mp4"/>
+      <pwccode url="https://github.com/edchengg/model_selection" additional="false">edchengg/model_selection</pwccode>
     </paper>
     <paper id="460">
       <title>Continual Few-Shot Learning for Text Classification</title>
@@ -7133,11 +7137,11 @@
       <url hash="504d4f84">2021.emnlp-main.460</url>
       <bibkey>pasunuru-etal-2021-continual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.460</doi>
+      <video href="2021.emnlp-main.460.mp4"/>
       <pwccode url="https://github.com/ramakanth-pasunuru/cfl-benchmark" additional="false">ramakanth-pasunuru/cfl-benchmark</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.460.mp4"/>
     </paper>
     <paper id="461">
       <title>Efficient Nearest Neighbor Language Models</title>
@@ -7149,9 +7153,10 @@
       <url hash="bbcbf48d">2021.emnlp-main.461</url>
       <bibkey>he-etal-2021-efficient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.461</doi>
+      <video href="2021.emnlp-main.461.mp4"/>
       <pwccode url="https://github.com/jxhe/efficient-knnlm" additional="false">jxhe/efficient-knnlm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.emnlp-main.461.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="462">
       <title><fixed-case>ST</fixed-case>ra<fixed-case>TA</fixed-case>: Self-Training with Task Augmentation for Better Few-shot Learning</title>
@@ -7165,13 +7170,13 @@
       <url hash="8f8c14aa">2021.emnlp-main.462</url>
       <bibkey>vu-etal-2021-strata</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.462</doi>
+      <video href="2021.emnlp-main.462.mp4"/>
       <pwccode url="https://github.com/google-research/google-research/tree/master/STraTA" additional="false">google-research/google-research</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.462.mp4"/>
     </paper>
     <paper id="463">
       <title><fixed-case>TADPOLE</fixed-case>: <fixed-case>T</fixed-case>ask <fixed-case>AD</fixed-case>apted <fixed-case>P</fixed-case>re-Training via <fixed-case>A</fixed-case>n<fixed-case>O</fixed-case>ma<fixed-case>L</fixed-case>y <fixed-case>D</fixed-case><fixed-case>E</fixed-case>tection</title>
@@ -7183,10 +7188,10 @@
       <url hash="d3bcac39">2021.emnlp-main.463</url>
       <bibkey>madan-etal-2021-tadpole</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.463</doi>
+      <video href="2021.emnlp-main.463.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
-      <video href="2021.emnlp-main.463.mp4"/>
     </paper>
     <paper id="464">
       <title>Gradient-based Adversarial Attacks against Text Transformers</title>
@@ -7199,10 +7204,10 @@
       <url hash="41d427f7">2021.emnlp-main.464</url>
       <bibkey>guo-etal-2021-gradient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.464</doi>
+      <video href="2021.emnlp-main.464.mp4"/>
       <pwccode url="https://github.com/facebookresearch/text-adversarial-attack" additional="false">facebookresearch/text-adversarial-attack</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.464.mp4"/>
     </paper>
     <paper id="465">
       <title>Do Transformer Modifications Transfer Across Implementations and Applications?</title>
@@ -7227,11 +7232,11 @@
       <url hash="f395c653">2021.emnlp-main.465</url>
       <bibkey>narang-etal-2021-transformer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.465</doi>
+      <video href="2021.emnlp-main.465.mp4"/>
       <pwccode url="https://github.com/google-research/google-research/tree/master/transformer_modifications/" additional="false">google-research/google-research</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
-      <video href="2021.emnlp-main.465.mp4"/>
     </paper>
     <paper id="466">
       <title>Paired Examples as Indirect Supervision in Latent Decision Models</title>
@@ -7260,12 +7265,12 @@
       <url hash="0eb1a54c">2021.emnlp-main.467</url>
       <bibkey>zhang-etal-2021-pairwise</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.467</doi>
+      <video href="2021.emnlp-main.467.mp4"/>
       <pwccode url="https://github.com/amazon-research/sentence-representations" additional="false">amazon-research/sentence-representations</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.emnlp-main.467.mp4"/>
     </paper>
     <paper id="468">
       <title>Muppet: Massive Multi-task Representations with Pre-Finetuning</title>
@@ -7280,6 +7285,7 @@
       <url hash="529197ff">2021.emnlp-main.468</url>
       <bibkey>aghajanyan-etal-2021-muppet</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.468</doi>
+      <video href="2021.emnlp-main.468.mp4"/>
       <pwccode url="https://huggingface.co/facebook/muppet-roberta-base" additional="true">facebook/muppet-roberta-base</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
@@ -7297,7 +7303,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/swag">SWAG</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.emnlp-main.468.mp4"/>
     </paper>
     <paper id="469">
       <title>Diverse Distributions of Self-Supervised Tasks for Meta-Learning in <fixed-case>NLP</fixed-case></title>
@@ -7311,11 +7316,11 @@
       <url hash="b816a7df">2021.emnlp-main.469</url>
       <bibkey>bansal-etal-2021-diverse</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.469</doi>
+      <video href="2021.emnlp-main.469.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ccnet">CCNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel-2-0">FewRel 2.0</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
-      <video href="2021.emnlp-main.469.mp4"/>
     </paper>
     <paper id="470">
       <title>A Simple and Effective Method To Eliminate the Self Language Bias in Multilingual Representations</title>
@@ -7328,10 +7333,10 @@
       <url hash="8efe8383">2021.emnlp-main.470</url>
       <bibkey>yang-etal-2021-simple</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.470</doi>
+      <video href="2021.emnlp-main.470.mp4"/>
       <pwccode url="https://github.com/ziyi-yang/lir" additional="false">ziyi-yang/lir</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad-r">LAReQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wiki-40b">Wiki-40B</pwcdataset>
-      <video href="2021.emnlp-main.470.mp4"/>
     </paper>
     <paper id="471">
       <title>A Massively Multilingual Analysis of Cross-linguality in Shared Embedding Space</title>
@@ -7344,8 +7349,8 @@
       <attachment type="Software" hash="6e5c2a8b">2021.emnlp-main.471.Software.zip</attachment>
       <bibkey>jones-etal-2021-massively</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.471</doi>
-      <pwccode url="https://github.com/alexjonesnlp/xlanalysis5k" additional="false">alexjonesnlp/xlanalysis5k</pwccode>
       <video href="2021.emnlp-main.471.mp4"/>
+      <pwccode url="https://github.com/alexjonesnlp/xlanalysis5k" additional="false">alexjonesnlp/xlanalysis5k</pwccode>
     </paper>
     <paper id="472">
       <title>Frustratingly Simple but Surprisingly Strong: Using Language-Independent Features for Zero-shot Cross-lingual Semantic Parsing</title>
@@ -7359,8 +7364,8 @@
       <attachment type="Software" hash="23f7f604">2021.emnlp-main.472.Software.zip</attachment>
       <bibkey>yang-etal-2021-frustratingly</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.472</doi>
-      <pwccode url="https://github.com/gt-salt/multilingual-drs-semantic-parsing" additional="false">gt-salt/multilingual-drs-semantic-parsing</pwccode>
       <video href="2021.emnlp-main.472.mp4"/>
+      <pwccode url="https://github.com/gt-salt/multilingual-drs-semantic-parsing" additional="false">gt-salt/multilingual-drs-semantic-parsing</pwccode>
     </paper>
     <paper id="473">
       <title>Improving Simultaneous Translation by Incorporating Pseudo-References with Fewer Reorderings</title>
@@ -7389,8 +7394,8 @@
       <url hash="db4e751d">2021.emnlp-main.474</url>
       <bibkey>sun-etal-2021-classification</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.474</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/mlqe-pe">MLQE-PE</pwcdataset>
       <video href="2021.emnlp-main.474.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mlqe-pe">MLQE-PE</pwcdataset>
     </paper>
     <paper id="475">
       <title>A Large-Scale Study of Machine Translation in <fixed-case>T</fixed-case>urkic Languages</title>
@@ -7415,8 +7420,8 @@
       <url hash="ac6ae7e8">2021.emnlp-main.475</url>
       <bibkey>mirzakhalov-etal-2021-large</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.475</doi>
-      <pwccode url="https://github.com/turkic-interlingua/til-mt" additional="false">turkic-interlingua/til-mt</pwccode>
       <video href="2021.emnlp-main.475.mp4"/>
+      <pwccode url="https://github.com/turkic-interlingua/til-mt" additional="false">turkic-interlingua/til-mt</pwccode>
     </paper>
     <paper id="476">
       <title>Analyzing the Surprising Variability in Word Embedding Stability Across Languages</title>
@@ -7428,8 +7433,8 @@
       <url hash="2331dae8">2021.emnlp-main.476</url>
       <bibkey>burdick-etal-2021-analyzing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.476</doi>
-      <pwccode url="https://github.com/laura-burdick/multilingual-stability" additional="false">laura-burdick/multilingual-stability</pwccode>
       <video href="2021.emnlp-main.476.mp4"/>
+      <pwccode url="https://github.com/laura-burdick/multilingual-stability" additional="false">laura-burdick/multilingual-stability</pwccode>
     </paper>
     <paper id="477">
       <title>Rule-based Morphological Inflection Improves Neural Terminology Translation</title>
@@ -7440,8 +7445,8 @@
       <url hash="e31fa80c">2021.emnlp-main.477</url>
       <bibkey>xu-carpuat-2021-rule</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.477</doi>
-      <pwccode url="https://github.com/izecson/terminology-translation" additional="false">izecson/terminology-translation</pwccode>
       <video href="2021.emnlp-main.477.mp4"/>
+      <pwccode url="https://github.com/izecson/terminology-translation" additional="false">izecson/terminology-translation</pwccode>
     </paper>
     <paper id="478">
       <title>Data and Parameter Scaling Laws for Neural Machine Translation</title>
@@ -7453,8 +7458,8 @@
       <url hash="8f42801d">2021.emnlp-main.478</url>
       <bibkey>gordon-etal-2021-data</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.478</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
       <video href="2021.emnlp-main.478.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
     </paper>
     <paper id="479">
       <title>Good-Enough Example Extrapolation</title>
@@ -7465,9 +7470,9 @@
       <attachment type="Software" hash="31382f34">2021.emnlp-main.479.Software.zip</attachment>
       <bibkey>wei-2021-good</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.479</doi>
+      <video href="2021.emnlp-main.479.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.emnlp-main.479.mp4"/>
     </paper>
     <paper id="480">
       <title>Learning to Selectively Learn for Weakly-supervised Paraphrase Generation</title>
@@ -7483,8 +7488,8 @@
       <url hash="1a9935ee">2021.emnlp-main.480</url>
       <bibkey>ding-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.480</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <video href="2021.emnlp-main.480.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
     </paper>
     <paper id="481">
       <title>Effective Convolutional Attention Network for Multi-label Clinical Document Classification</title>
@@ -7513,9 +7518,9 @@
       <url hash="3d293e81">2021.emnlp-main.482</url>
       <bibkey>jain-etal-2021-contrastive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.482</doi>
+      <video href="2021.emnlp-main.482.mp4"/>
       <pwccode url="https://github.com/parasj/contracode" additional="false">parasj/contracode</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/codesearchnet">CodeSearchNet</pwcdataset>
-      <video href="2021.emnlp-main.482.mp4"/>
     </paper>
     <paper id="483">
       <title><fixed-case>IGA</fixed-case>: An Intent-Guided Authoring Assistant</title>
@@ -7532,9 +7537,9 @@
       <url hash="bccb34f3">2021.emnlp-main.483</url>
       <bibkey>sun-etal-2021-iga</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.483</doi>
+      <video href="2021.emnlp-main.483.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/pomo">PoMo</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikilarge">WikiLarge</pwcdataset>
-      <video href="2021.emnlp-main.483.mp4"/>
     </paper>
     <paper id="484">
       <title>Math Word Problem Generation with Mathematical Consistency and Problem Context Constraints</title>
@@ -7546,9 +7551,9 @@
       <url hash="c7ab2234">2021.emnlp-main.484</url>
       <bibkey>wang-etal-2021-math</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.484</doi>
+      <video href="2021.emnlp-main.484.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/math23k">Math23K</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mathqa">MathQA</pwcdataset>
-      <video href="2021.emnlp-main.484.mp4"/>
     </paper>
     <paper id="485">
       <title>Navigating the Kaleidoscope of <fixed-case>COVID</fixed-case>-19 Misinformation Using Deep Learning</title>
@@ -7571,8 +7576,8 @@
       <url hash="0a65c1ff">2021.emnlp-main.486</url>
       <bibkey>li-etal-2021-detecting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.486</doi>
-      <pwccode url="https://github.com/junwang4/detecting-health-advice" additional="false">junwang4/detecting-health-advice</pwccode>
       <video href="2021.emnlp-main.486.mp4"/>
+      <pwccode url="https://github.com/junwang4/detecting-health-advice" additional="false">junwang4/detecting-health-advice</pwccode>
     </paper>
     <paper id="487">
       <title>A Semantic Feature-Wise Transformation Relation Network for Automatic Short Answer Grading</title>
@@ -7595,9 +7600,9 @@
       <url hash="ab13dd8b">2021.emnlp-main.488</url>
       <bibkey>manchanda-karypis-2021-evaluating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.488</doi>
+      <video href="2021.emnlp-main.488.mp4"/>
       <pwccode url="https://github.com/gurdaspuriya/evaluating-scholarly-impact" additional="false">gurdaspuriya/evaluating-scholarly-impact</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/s2orc">S2ORC</pwcdataset>
-      <video href="2021.emnlp-main.488.mp4"/>
     </paper>
     <paper id="489">
       <title>A Scalable Framework for Learning From Implicit User Feedback to Improve Natural Language Understanding in Large-Scale Conversational <fixed-case>AI</fixed-case> Systems</title>
@@ -7628,10 +7633,10 @@
       <url hash="6ae2e8d7">2021.emnlp-main.490</url>
       <bibkey>inoue-etal-2021-summarize</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.490</doi>
+      <video href="2021.emnlp-main.490.mp4"/>
       <pwccode url="https://github.com/stonybrooknlp/suqa" additional="false">stonybrooknlp/suqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
-      <video href="2021.emnlp-main.490.mp4"/>
     </paper>
     <paper id="491">
       <title><fixed-case>F</fixed-case>ewshot<fixed-case>QA</fixed-case>: A simple framework for few-shot learning of question answering tasks using pre-trained text-to-text models</title>
@@ -7642,10 +7647,10 @@
       <url hash="4272c2d9">2021.emnlp-main.491</url>
       <bibkey>chada-natarajan-2021-fewshotqa</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.491</doi>
+      <video href="2021.emnlp-main.491.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/mrqa-2019">MRQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
-      <video href="2021.emnlp-main.491.mp4"/>
     </paper>
     <paper id="492">
       <title>Multi-stage Training with Improved Negative Contrast for Neural Passage Retrieval</title>
@@ -7659,9 +7664,9 @@
       <url hash="e4443ebb">2021.emnlp-main.492</url>
       <bibkey>lu-etal-2021-multi</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.492</doi>
+      <video href="2021.emnlp-main.492.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.emnlp-main.492.mp4"/>
     </paper>
     <paper id="493">
       <title>Perhaps <fixed-case>PTLM</fixed-case>s Should Go to School – A Task to Assess Open Book and Closed Book <fixed-case>QA</fixed-case></title>
@@ -7676,10 +7681,10 @@
       <url hash="586e0f3f">2021.emnlp-main.493</url>
       <bibkey>ciosici-etal-2021-perhaps</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.493</doi>
+      <video href="2021.emnlp-main.493.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/arc">ARC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.493.mp4"/>
     </paper>
     <paper id="494">
       <title><fixed-case>R</fixed-case>eason<fixed-case>BERT</fixed-case>: <fixed-case>P</fixed-case>re-trained to Reason with Distant Supervision</title>
@@ -7694,6 +7699,7 @@
       <url hash="8bdf3941">2021.emnlp-main.494</url>
       <bibkey>deng-etal-2021-reasonbert</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.494</doi>
+      <video href="2021.emnlp-main.494.mp4"/>
       <pwccode url="https://github.com/sunlab-osu/reasonbert" additional="false">sunlab-osu/reasonbert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hybridqa">HybridQA</pwcdataset>
@@ -7703,7 +7709,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/searchqa">SearchQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.emnlp-main.494.mp4"/>
     </paper>
     <paper id="495">
       <title>Single-dataset Experts for Multi-dataset Question Answering</title>
@@ -7715,6 +7720,7 @@
       <url hash="dc747808">2021.emnlp-main.495</url>
       <bibkey>friedman-etal-2021-single</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.495</doi>
+      <video href="2021.emnlp-main.495.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/made" additional="false">princeton-nlp/made</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/duorc">DuoRC</pwcdataset>
@@ -7724,7 +7730,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.emnlp-main.495.mp4"/>
     </paper>
     <paper id="496">
       <title>Simple Entity-Centric Questions Challenge Dense Retrievers</title>
@@ -7737,11 +7742,11 @@
       <url hash="3a787a5a">2021.emnlp-main.496</url>
       <bibkey>sciavolino-etal-2021-simple</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.496</doi>
+      <video href="2021.emnlp-main.496.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/entityquestions" additional="false">princeton-nlp/entityquestions</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/entityquestions">EntityQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paq">PAQ</pwcdataset>
-      <video href="2021.emnlp-main.496.mp4"/>
     </paper>
     <paper id="497">
       <title>Mitigating False-Negative Contexts in Multi-document Question Answering with Retrieval Marginalization</title>
@@ -7753,11 +7758,11 @@
       <url hash="c6f18453">2021.emnlp-main.497</url>
       <bibkey>ni-etal-2021-mitigating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.497</doi>
+      <video href="2021.emnlp-main.497.mp4"/>
       <pwccode url="https://github.com/niansong1996/retrieval_marginalization" additional="false">niansong1996/retrieval_marginalization</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iirc">IIRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.emnlp-main.497.mp4"/>
     </paper>
     <paper id="498">
       <title><fixed-case>M</fixed-case>ulti<fixed-case>D</fixed-case>oc2<fixed-case>D</fixed-case>ial: Modeling Dialogues Grounded in Multiple Documents</title>
@@ -7770,6 +7775,7 @@
       <url hash="24beff46">2021.emnlp-main.498</url>
       <bibkey>feng-etal-2021-multidoc2dial</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.498</doi>
+      <video href="2021.emnlp-main.498.mp4"/>
       <pwccode url="https://github.com/IBM/multidoc2dial" additional="false">IBM/multidoc2dial</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multidoc2dial">MultiDoc2Dial</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/doqa">DoQA</pwcdataset>
@@ -7778,7 +7784,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/quac">QuAC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sharc">ShARC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/doc2dial">doc2dial</pwcdataset>
-      <video href="2021.emnlp-main.498.mp4"/>
     </paper>
     <paper id="499">
       <title><fixed-case>G</fixed-case>up<fixed-case>S</fixed-case>hup: Summarizing Open-Domain Code-Switched Conversations</title>
@@ -7797,9 +7802,9 @@
       <url hash="712d3676">2021.emnlp-main.499</url>
       <bibkey>mehnaz-etal-2021-gupshup</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.499</doi>
+      <video href="2021.emnlp-main.499.mp4"/>
       <pwccode url="https://github.com/midas-research/gupshup" additional="false">midas-research/gupshup</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.emnlp-main.499.mp4"/>
     </paper>
     <paper id="500">
       <title><fixed-case>B</fixed-case>i<fixed-case>SECT</fixed-case>: Learning to Split and Rephrase Sentences with Bitexts</title>
@@ -7813,9 +7818,9 @@
       <url hash="6affb1a8">2021.emnlp-main.500</url>
       <bibkey>kim-etal-2021-bisect</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.500</doi>
+      <video href="2021.emnlp-main.500.mp4"/>
       <pwccode url="https://github.com/mounicam/bisect" additional="false">mounicam/bisect</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bisect">BiSECT</pwcdataset>
-      <video href="2021.emnlp-main.500.mp4"/>
     </paper>
     <paper id="501">
       <title>Data Collection vs. Knowledge Graph Completion: What is Needed to Improve Coverage?</title>
@@ -7840,10 +7845,10 @@
       <url hash="706ceda5">2021.emnlp-main.502</url>
       <bibkey>yang-etal-2021-universal</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.502</doi>
+      <video href="2021.emnlp-main.502.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.emnlp-main.502.mp4"/>
     </paper>
     <paper id="503">
       <title>On the Benefit of Syntactic Supervision for Cross-lingual Transfer in Semantic Role Labeling</title>
@@ -7855,9 +7860,9 @@
       <url hash="b4124564">2021.emnlp-main.503</url>
       <bibkey>zhang-etal-2021-benefit</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.503</doi>
+      <video href="2021.emnlp-main.503.mp4"/>
       <pwccode url="https://github.com/zzsfornlp/zmsp" additional="false">zzsfornlp/zmsp</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.emnlp-main.503.mp4"/>
     </paper>
     <paper id="504">
       <title>Implicit Premise Generation with Discourse-aware Commonsense Knowledge Models</title>
@@ -7869,9 +7874,9 @@
       <url hash="5dbdab40">2021.emnlp-main.504</url>
       <bibkey>chakrabarty-etal-2021-implicit</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.504</doi>
+      <video href="2021.emnlp-main.504.mp4"/>
       <pwccode url="https://github.com/tuhinjubcse/enthymemesemnlp2021" additional="false">tuhinjubcse/enthymemesemnlp2021</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/art-dataset">ART Dataset</pwcdataset>
-      <video href="2021.emnlp-main.504.mp4"/>
     </paper>
     <paper id="505">
       <title>Inducing Transformer’s Compositional Generalization Ability via Auxiliary Sequence Prediction Tasks</title>
@@ -7882,10 +7887,10 @@
       <url hash="eb65d8de">2021.emnlp-main.505</url>
       <bibkey>jiang-bansal-2021-inducing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.505</doi>
+      <video href="2021.emnlp-main.505.mp4"/>
       <pwccode url="https://github.com/jiangyctarheel/compositional-auxseq" additional="false">jiangyctarheel/compositional-auxseq</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
-      <video href="2021.emnlp-main.505.mp4"/>
     </paper>
     <paper id="506">
       <title>Flexible Generation of Natural Language Deductions</title>
@@ -7899,11 +7904,11 @@
       <attachment type="Software" hash="05f29bbe">2021.emnlp-main.506.Software.zip</attachment>
       <bibkey>bostrom-etal-2021-flexible</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.506</doi>
+      <video href="2021.emnlp-main.506.mp4"/>
       <pwccode url="https://github.com/alephic/ParaPattern" additional="false">alephic/ParaPattern</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/entailmentbank">EntailmentBank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/proofwriter">ProofWriter</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qasc">QASC</pwcdataset>
-      <video href="2021.emnlp-main.506.mp4"/>
     </paper>
     <paper id="507">
       <title>Structure-aware Fine-tuning of Sequence-to-sequence Transformers for Transition-based <fixed-case>AMR</fixed-case> Parsing</title>
@@ -7918,8 +7923,8 @@
       <url hash="0cfd512f">2021.emnlp-main.507</url>
       <bibkey>zhou-etal-2021-structure</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.507</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/ldc2017t10">LDC2017T10</pwcdataset>
       <video href="2021.emnlp-main.507.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ldc2017t10">LDC2017T10</pwcdataset>
     </paper>
     <paper id="508">
       <title>Think about it! Improving defeasible reasoning by first modeling the question scenario.</title>
@@ -7934,11 +7939,11 @@
       <url hash="9b25d655">2021.emnlp-main.508</url>
       <bibkey>madaan-etal-2021-think</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.508</doi>
+      <video href="2021.emnlp-main.508.mp4"/>
       <pwccode url="https://github.com/madaan/thinkaboutit" additional="false">madaan/thinkaboutit</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/atomic">ATOMIC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wiqa">WIQA</pwcdataset>
-      <video href="2021.emnlp-main.508.mp4"/>
     </paper>
     <paper id="509">
       <title>Open Aspect Target Sentiment Classification with Natural Language Prompts</title>
@@ -7953,9 +7958,9 @@
       <url hash="925372c6">2021.emnlp-main.509</url>
       <bibkey>seoh-etal-2021-open</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.509</doi>
+      <video href="2021.emnlp-main.509.mp4"/>
       <pwccode url="https://github.com/ronaldseoh/atsc_prompts" additional="false">ronaldseoh/atsc_prompts</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/semeval-2014-task-4-sub-task-2">SemEval 2014 Task 4 Sub Task 2</pwcdataset>
-      <video href="2021.emnlp-main.509.mp4"/>
     </paper>
     <paper id="510">
       <title>Does <fixed-case>BERT</fixed-case> Learn as Humans Perceive? Understanding Linguistic Styles through Lexica</title>
@@ -7967,9 +7972,9 @@
       <url hash="068a99d5">2021.emnlp-main.510</url>
       <bibkey>hayati-etal-2021-bert</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.510</doi>
+      <video href="2021.emnlp-main.510.mp4"/>
       <pwccode url="https://github.com/sweetpeach/hummingbird" additional="false">sweetpeach/hummingbird</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hummingbird">Hummingbird</pwcdataset>
-      <video href="2021.emnlp-main.510.mp4"/>
     </paper>
     <paper id="511">
       <title>Improving Stance Detection with Multi-Dataset Learning and Knowledge Distillation</title>
@@ -7981,8 +7986,8 @@
       <url hash="0bb31b69">2021.emnlp-main.511</url>
       <bibkey>li-etal-2021-improving-stance</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.511</doi>
-      <pwccode url="https://github.com/chuchun8/mdl-stance-distillation" additional="false">chuchun8/mdl-stance-distillation</pwccode>
       <video href="2021.emnlp-main.511.mp4"/>
+      <pwccode url="https://github.com/chuchun8/mdl-stance-distillation" additional="false">chuchun8/mdl-stance-distillation</pwccode>
     </paper>
     <paper id="512">
       <title>Discovering the Unknown Knowns: Turning Implicit Knowledge in the Dataset into Explicit Training Examples for Visual Question Answering</title>
@@ -7995,12 +8000,12 @@
       <url hash="63c28a40">2021.emnlp-main.512</url>
       <bibkey>kil-etal-2021-discovering</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.512</doi>
+      <video href="2021.emnlp-main.512.mp4"/>
       <pwccode url="https://github.com/heendung/simpleaug" additional="false">heendung/simpleaug</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering-v2-0">Visual Question Answering v2.0</pwcdataset>
-      <video href="2021.emnlp-main.512.mp4"/>
     </paper>
     <paper id="513">
       <title>Improving Pre-trained Vision-and-Language Embeddings for Phrase Grounding</title>
@@ -8011,9 +8016,9 @@
       <url hash="f851e2f4">2021.emnlp-main.513</url>
       <bibkey>dou-peng-2021-improving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.513</doi>
+      <video href="2021.emnlp-main.513.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.emnlp-main.513.mp4"/>
     </paper>
     <paper id="514">
       <title>Sequential Randomized Smoothing for Adversarially Robust Speech Recognition</title>
@@ -8024,9 +8029,9 @@
       <url hash="c4007561">2021.emnlp-main.514</url>
       <bibkey>olivier-raj-2021-sequential</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.514</doi>
+      <video href="2021.emnlp-main.514.mp4"/>
       <pwccode url="https://github.com/raphaelolivier/smoothingasr" additional="false">raphaelolivier/smoothingasr</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
-      <video href="2021.emnlp-main.514.mp4"/>
     </paper>
     <paper id="515">
       <title>Hitting your <fixed-case>MARQ</fixed-case>: Multimodal <fixed-case>AR</fixed-case>gument Quality Assessment in Long Debate Video</title>
@@ -8042,9 +8047,9 @@
       <url hash="e8e5cbb2">2021.emnlp-main.515</url>
       <bibkey>hasan-etal-2021-hitting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.515</doi>
+      <video href="2021.emnlp-main.515.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dbates">DBATES</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ibm-rank-30k">IBM-Rank-30k</pwcdataset>
-      <video href="2021.emnlp-main.515.mp4"/>
     </paper>
     <paper id="516">
       <title>Mind the Context: The Impact of Contextualization in Neural Module Networks for Grounding Visual Referring Expressions</title>
@@ -8058,9 +8063,9 @@
       <url hash="8f7d835d">2021.emnlp-main.516</url>
       <bibkey>akula-etal-2021-mind</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.516</doi>
+      <video href="2021.emnlp-main.516.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/clevr">CLEVR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/clevr-ref">CLEVR-Ref+</pwcdataset>
-      <video href="2021.emnlp-main.516.mp4"/>
     </paper>
     <paper id="517">
       <title>Weakly-Supervised Visual-Retriever-Reader for Knowledge-based Question Answering</title>
@@ -8073,11 +8078,11 @@
       <url hash="f0515533">2021.emnlp-main.517</url>
       <bibkey>luo-etal-2021-weakly</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.517</doi>
+      <video href="2021.emnlp-main.517.mp4"/>
       <pwccode url="https://github.com/luomancs/retriever_reader_for_okvqa" additional="false">luomancs/retriever_reader_for_okvqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ok-vqa">OK-VQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.emnlp-main.517.mp4"/>
     </paper>
     <paper id="518">
       <title><fixed-case>NDH</fixed-case>-Full: Learning and Evaluating Navigational Agents on Full-Length Dialogue</title>
@@ -8089,8 +8094,8 @@
       <url hash="55ecd0ef">2021.emnlp-main.518</url>
       <bibkey>kim-etal-2021-ndh</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.518</doi>
-      <pwccode url="https://github.com/hyounghk/ndh-full" additional="false">hyounghk/ndh-full</pwccode>
       <video href="2021.emnlp-main.518.mp4"/>
+      <pwccode url="https://github.com/hyounghk/ndh-full" additional="false">hyounghk/ndh-full</pwccode>
     </paper>
     <paper id="519">
       <title>Timeline Summarization based on Event Graph Compression via Time-Aware Optimal Transport</title>
@@ -8106,8 +8111,8 @@
       <url hash="22b6177e">2021.emnlp-main.519</url>
       <bibkey>li-etal-2021-timeline</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.519</doi>
-      <pwccode url="https://github.com/limanling/event-graph-summarization" additional="false">limanling/event-graph-summarization</pwccode>
       <video href="2021.emnlp-main.519.mp4"/>
+      <pwccode url="https://github.com/limanling/event-graph-summarization" additional="false">limanling/event-graph-summarization</pwccode>
     </paper>
     <paper id="520">
       <title><fixed-case>S</fixed-case>tream<fixed-case>H</fixed-case>over: Livestream Transcript Summarization and Annotation</title>
@@ -8126,8 +8131,8 @@
       <url hash="d9e4e71f">2021.emnlp-main.520</url>
       <bibkey>cho-etal-2021-streamhover</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.520</doi>
-      <pwccode url="https://github.com/ucfnlp/streamhover" additional="false">ucfnlp/streamhover</pwccode>
       <video href="2021.emnlp-main.520.mp4"/>
+      <pwccode url="https://github.com/ucfnlp/streamhover" additional="false">ucfnlp/streamhover</pwccode>
     </paper>
     <paper id="521">
       <title>Cross-Register Projection for Headline Part of Speech Tagging</title>
@@ -8139,9 +8144,9 @@
       <url hash="3210a2d1">2021.emnlp-main.521</url>
       <bibkey>benton-etal-2021-cross</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.521</doi>
+      <video href="2021.emnlp-main.521.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/english-web-treebank">English Web Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sentence-compression">Sentence Compression</pwcdataset>
-      <video href="2021.emnlp-main.521.mp4"/>
     </paper>
     <paper id="522">
       <title>Editing Factual Knowledge in Language Models</title>
@@ -8165,10 +8170,10 @@
       <url hash="543a4f27">2021.emnlp-main.523</url>
       <bibkey>zhang-etal-2021-sparse</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.523</doi>
+      <video href="2021.emnlp-main.523.mp4"/>
       <pwccode url="https://github.com/bzhangGo/zero" additional="true">bzhangGo/zero</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2016">WMT 2016</pwcdataset>
-      <video href="2021.emnlp-main.523.mp4"/>
     </paper>
     <paper id="524">
       <title>Knowledge Base Completion Meets Transfer Learning</title>
@@ -8179,9 +8184,9 @@
       <url hash="9a2550f8">2021.emnlp-main.524</url>
       <bibkey>kocijan-lukasiewicz-2021-knowledge</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.524</doi>
+      <video href="2021.emnlp-main.524.mp4"/>
       <pwccode url="https://github.com/vid-koci/kbctransferlearning" additional="false">vid-koci/kbctransferlearning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/olpbench">OLPBENCH</pwcdataset>
-      <video href="2021.emnlp-main.524.mp4"/>
     </paper>
     <paper id="525">
       <title><fixed-case>SPECTRA</fixed-case>: Sparse Structured Text Rationalization</title>
@@ -8192,11 +8197,11 @@
       <url hash="e0ef9dc5">2021.emnlp-main.525</url>
       <bibkey>guerreiro-martins-2021-spectra</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.525</doi>
+      <video href="2021.emnlp-main.525.mp4"/>
       <pwccode url="https://github.com/deep-spin/spectra-rationalization" additional="true">deep-spin/spectra-rationalization</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.525.mp4"/>
     </paper>
     <paper id="526">
       <title>Towards Zero-Shot Knowledge Distillation for Natural Language Processing</title>
@@ -8209,11 +8214,11 @@
       <url hash="01c17615">2021.emnlp-main.526</url>
       <bibkey>rashid-etal-2021-towards</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.526</doi>
+      <video href="2021.emnlp-main.526.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.526.mp4"/>
     </paper>
     <paper id="527">
       <title>Adversarial Regularization as Stackelberg Game: An Unrolled Optimization Approach</title>
@@ -8230,13 +8235,13 @@
       <url hash="13bdb63c">2021.emnlp-main.527</url>
       <bibkey>zuo-etal-2021-adversarial</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.527</doi>
+      <video href="2021.emnlp-main.527.mp4"/>
       <pwccode url="https://github.com/SimiaoZuo/Stackelberg-Adv" additional="false">SimiaoZuo/Stackelberg-Adv</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.527.mp4"/>
     </paper>
     <paper id="528">
       <title>Aspect-Controllable Opinion Summarization</title>
@@ -8248,8 +8253,8 @@
       <url hash="e65dc561">2021.emnlp-main.528</url>
       <bibkey>amplayo-etal-2021-aspect</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.528</doi>
-      <pwccode url="https://github.com/rktamplayo/acesum" additional="false">rktamplayo/acesum</pwccode>
       <video href="2021.emnlp-main.528.mp4"/>
+      <pwccode url="https://github.com/rktamplayo/acesum" additional="false">rktamplayo/acesum</pwccode>
     </paper>
     <paper id="529">
       <title><fixed-case>Q</fixed-case>uest<fixed-case>E</fixed-case>val: Summarization Asks for Fact-based Evaluation</title>
@@ -8265,8 +8270,8 @@
       <url hash="cf58cfa9">2021.emnlp-main.529</url>
       <bibkey>scialom-etal-2021-questeval</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.529</doi>
-      <pwccode url="https://github.com/recitalAI/QuestEval" additional="true">recitalAI/QuestEval</pwccode>
       <video href="2021.emnlp-main.529.mp4"/>
+      <pwccode url="https://github.com/recitalAI/QuestEval" additional="true">recitalAI/QuestEval</pwccode>
     </paper>
     <paper id="530">
       <title>Simple Conversational Data Augmentation for Semi-supervised Abstractive Dialogue Summarization</title>
@@ -8277,9 +8282,9 @@
       <url hash="62be795e">2021.emnlp-main.530</url>
       <bibkey>chen-yang-2021-simple</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.530</doi>
+      <video href="2021.emnlp-main.530.mp4"/>
       <pwccode url="https://github.com/gt-salt/coda" additional="false">gt-salt/coda</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.emnlp-main.530.mp4"/>
     </paper>
     <paper id="531">
       <title>Finding a Balanced Degree of Automation for Summary Evaluation</title>
@@ -8290,8 +8295,8 @@
       <url hash="071dc2a3">2021.emnlp-main.531</url>
       <bibkey>zhang-bansal-2021-finding</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.531</doi>
-      <pwccode url="https://github.com/zhangshiyue/lite2-3pyramid" additional="false">zhangshiyue/lite2-3pyramid</pwccode>
       <video href="2021.emnlp-main.531.mp4"/>
+      <pwccode url="https://github.com/zhangshiyue/lite2-3pyramid" additional="false">zhangshiyue/lite2-3pyramid</pwccode>
     </paper>
     <paper id="532">
       <title><fixed-case>CLIFF</fixed-case>: Contrastive Learning for Improving Faithfulness and Factuality in Abstractive Summarization</title>
@@ -8302,8 +8307,8 @@
       <url hash="7337a237">2021.emnlp-main.532</url>
       <bibkey>cao-wang-2021-cliff</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.532</doi>
-      <pwccode url="https://github.com/makcedward/nlpaug" additional="false">makcedward/nlpaug</pwccode>
       <video href="2021.emnlp-main.532.mp4"/>
+      <pwccode url="https://github.com/makcedward/nlpaug" additional="false">makcedward/nlpaug</pwccode>
     </paper>
     <paper id="533">
       <title>Multilingual Unsupervised Neural Machine Translation with Denoising Adapters</title>
@@ -8316,8 +8321,8 @@
       <url hash="d983a648">2021.emnlp-main.533</url>
       <bibkey>ustun-etal-2021-multilingual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.533</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/flores">FLoRes</pwcdataset>
       <video href="2021.emnlp-main.533.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/flores">FLoRes</pwcdataset>
     </paper>
     <paper id="534">
       <title><fixed-case>BERT</fixed-case>, m<fixed-case>BERT</fixed-case>, or <fixed-case>B</fixed-case>i<fixed-case>BERT</fixed-case>? A Study on Contextualized Embeddings for Neural Machine Translation</title>
@@ -8329,10 +8334,10 @@
       <url hash="8d079ab2">2021.emnlp-main.534</url>
       <bibkey>xu-etal-2021-bert</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.534</doi>
+      <video href="2021.emnlp-main.534.mp4"/>
       <pwccode url="https://github.com/fe1ixxu/BiBERT" additional="true">fe1ixxu/BiBERT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/oscar">OSCAR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
-      <video href="2021.emnlp-main.534.mp4"/>
     </paper>
     <paper id="535">
       <title>Controlling Machine Translation for Multiple Attributes with Additive Interventions</title>
@@ -8345,8 +8350,8 @@
       <url hash="9e005586">2021.emnlp-main.535</url>
       <bibkey>schioppa-etal-2021-controlling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.535</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/jesc">JESC</pwcdataset>
       <video href="2021.emnlp-main.535.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/jesc">JESC</pwcdataset>
     </paper>
     <paper id="536">
       <title>A Generative Framework for Simultaneous Machine Translation</title>
@@ -8372,8 +8377,8 @@
       <url hash="a21c618c">2021.emnlp-main.537</url>
       <bibkey>zhao-etal-2021-good</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.537</doi>
-      <pwccode url="https://github.com/mingzi151/interpretationdata" additional="false">mingzi151/interpretationdata</pwccode>
       <video href="2021.emnlp-main.537.mp4"/>
+      <pwccode url="https://github.com/mingzi151/interpretationdata" additional="false">mingzi151/interpretationdata</pwccode>
     </paper>
     <paper id="538">
       <title>Boosting Cross-Lingual Transfer via Self-Learning with Uncertainty Estimation</title>
@@ -8388,9 +8393,9 @@
       <url hash="9dd570d4">2021.emnlp-main.538</url>
       <bibkey>xu-etal-2021-boosting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.538</doi>
+      <video href="2021.emnlp-main.538.mp4"/>
       <pwccode url="https://github.com/lxucs/multilingual-sl" additional="false">lxucs/multilingual-sl</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.emnlp-main.538.mp4"/>
     </paper>
     <paper id="539">
       <title><fixed-case>L</fixed-case>evenshtein Training for Word-level Quality Estimation</title>
@@ -8403,8 +8408,8 @@
       <url hash="a889fd4e">2021.emnlp-main.539</url>
       <bibkey>ding-etal-2021-levenshtein</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.539</doi>
-      <pwccode url="https://github.com/shuoyangd/stenella" additional="false">shuoyangd/stenella</pwccode>
       <video href="2021.emnlp-main.539.mp4"/>
+      <pwccode url="https://github.com/shuoyangd/stenella" additional="false">shuoyangd/stenella</pwccode>
     </paper>
     <paper id="540">
       <title>Interactive Machine Comprehension with Dynamic Knowledge Graphs</title>
@@ -8414,9 +8419,9 @@
       <url hash="0a7f6862">2021.emnlp-main.540</url>
       <bibkey>yuan-2021-interactive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.540</doi>
+      <video href="2021.emnlp-main.540.mp4"/>
       <pwccode url="https://github.com/xingdi-eric-yuan/imrc_graph_public" additional="false">xingdi-eric-yuan/imrc_graph_public</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.540.mp4"/>
     </paper>
     <paper id="541">
       <title>Residual Adapters for Parameter-Efficient <fixed-case>ASR</fixed-case> Adaptation to Atypical and Accented Speech</title>
@@ -8443,8 +8448,8 @@
       <url hash="fea3b545">2021.emnlp-main.542</url>
       <bibkey>liu-etal-2021-visual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.542</doi>
-      <pwccode url="https://github.com/FuxiaoLiu/VisualNews-Repository" additional="false">FuxiaoLiu/VisualNews-Repository</pwccode>
       <video href="2021.emnlp-main.542.mp4"/>
+      <pwccode url="https://github.com/FuxiaoLiu/VisualNews-Repository" additional="false">FuxiaoLiu/VisualNews-Repository</pwccode>
     </paper>
     <paper id="543">
       <title>Integrating Visuospatial, Linguistic, and Commonsense Structure into Story Visualization</title>
@@ -8455,10 +8460,10 @@
       <url hash="7e74d4b0">2021.emnlp-main.543</url>
       <bibkey>maharana-bansal-2021-integrating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.543</doi>
+      <video href="2021.emnlp-main.543.mp4"/>
       <pwccode url="https://github.com/adymaharana/vlcstorygan" additional="false">adymaharana/vlcstorygan</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
-      <video href="2021.emnlp-main.543.mp4"/>
     </paper>
     <paper id="544">
       <title><fixed-case>V</fixed-case>ideo<fixed-case>CLIP</fixed-case>: Contrastive Pre-training for Zero-shot Video-Text Understanding</title>
@@ -8475,13 +8480,14 @@
       <url hash="85cab672">2021.emnlp-main.544</url>
       <bibkey>xu-etal-2021-videoclip</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.544</doi>
+      <video href="2021.emnlp-main.544.mp4"/>
       <pwccode url="https://github.com/pytorch/fairseq" additional="true">pytorch/fairseq</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coin">COIN</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/crosstask">CrossTask</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/didemo">DiDeMo</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/howto100m">HowTo100M</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/msr-vtt">MSR-VTT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/youcook2">YouCook2</pwcdataset>
-      <video href="2021.emnlp-main.544.mp4"/>
     </paper>
     <paper id="545">
       <title><fixed-case>N</fixed-case>ews<fixed-case>CLIP</fixed-case>pings: <fixed-case>A</fixed-case>utomatic <fixed-case>G</fixed-case>eneration of <fixed-case>O</fixed-case>ut-of-<fixed-case>C</fixed-case>ontext <fixed-case>M</fixed-case>ultimodal <fixed-case>M</fixed-case>edia</title>
@@ -8493,10 +8499,10 @@
       <url hash="3b57d479">2021.emnlp-main.545</url>
       <bibkey>luo-etal-2021-newsclippings</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.545</doi>
+      <video href="2021.emnlp-main.545.mp4"/>
       <pwccode url="https://github.com/g-luo/news_clippings" additional="false">g-luo/news_clippings</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/newsclippings">NewsCLIPpings</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
-      <video href="2021.emnlp-main.545.mp4"/>
     </paper>
     <paper id="546">
       <title>Powering Comparative Classification with Sentiment Analysis via Domain Adaptive Knowledge Transfer</title>
@@ -8509,8 +8515,8 @@
       <url hash="5efbcdc0">2021.emnlp-main.546</url>
       <bibkey>li-etal-2021-powering</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.546</doi>
-      <pwccode url="https://github.com/zyli93/saecon" additional="false">zyli93/saecon</pwccode>
       <video href="2021.emnlp-main.546.mp4"/>
+      <pwccode url="https://github.com/zyli93/saecon" additional="false">zyli93/saecon</pwccode>
     </paper>
     <paper id="547">
       <title>Tribrid: Stance Classification with Neural Inconsistency Detection</title>
@@ -8522,9 +8528,9 @@
       <attachment type="Software" hash="42d7ad82">2021.emnlp-main.547.Software.zip</attachment>
       <bibkey>yang-urbani-2021-tribrid</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.547</doi>
+      <video href="2021.emnlp-main.547.mp4"/>
       <pwccode url="https://github.com/karmaresearch/tribrid" additional="false">karmaresearch/tribrid</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/perspectrum">Perspectrum</pwcdataset>
-      <video href="2021.emnlp-main.547.mp4"/>
     </paper>
     <paper id="548">
       <title><fixed-case>SYSML</fixed-case>: <fixed-case>S</fixed-case>t<fixed-case>Y</fixed-case>lometry with <fixed-case>S</fixed-case>tructure and <fixed-case>M</fixed-case>ultitask <fixed-case>L</fixed-case>earning: <fixed-case>I</fixed-case>mplications for <fixed-case>D</fixed-case>arknet Forum Migrant Analysis</title>
@@ -8536,8 +8542,8 @@
       <url hash="c3ac7c97">2021.emnlp-main.548</url>
       <bibkey>maneriker-etal-2021-sysml</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.548</doi>
-      <pwccode url="https://github.com/pranavmaneriker/sysml" additional="false">pranavmaneriker/sysml</pwccode>
       <video href="2021.emnlp-main.548.mp4"/>
+      <pwccode url="https://github.com/pranavmaneriker/sysml" additional="false">pranavmaneriker/sysml</pwccode>
     </paper>
     <paper id="549">
       <title>Few-Shot Emotion Recognition in Conversation with Sequential Prototypical Networks</title>
@@ -8551,9 +8557,9 @@
       <url hash="29d970cb">2021.emnlp-main.549</url>
       <bibkey>guibon-etal-2021-shot</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.549</doi>
+      <video href="2021.emnlp-main.549.mp4"/>
       <pwccode url="https://github.com/gguibon/protoseq" additional="false">gguibon/protoseq</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
-      <video href="2021.emnlp-main.549.mp4"/>
     </paper>
     <paper id="550">
       <title><fixed-case>CLASSIC</fixed-case>: Continual and Contrastive Learning of Aspect Sentiment Classification Tasks</title>
@@ -8566,8 +8572,8 @@
       <url hash="a9de4091">2021.emnlp-main.550</url>
       <bibkey>ke-etal-2021-classic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.550</doi>
-      <pwccode url="https://github.com/zixuanke/pycontinual" additional="false">zixuanke/pycontinual</pwccode>
       <video href="2021.emnlp-main.550.mp4"/>
+      <pwccode url="https://github.com/zixuanke/pycontinual" additional="false">zixuanke/pycontinual</pwccode>
     </paper>
     <paper id="551">
       <title>Implicit Sentiment Analysis with Event-centered Text Representation</title>
@@ -8580,8 +8586,8 @@
       <url hash="abef830f">2021.emnlp-main.551</url>
       <bibkey>zhou-etal-2021-implicit</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.551</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
       <video href="2021.emnlp-main.551.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
     </paper>
     <paper id="552">
       <title><fixed-case>S</fixed-case>im<fixed-case>CSE</fixed-case>: Simple Contrastive Learning of Sentence Embeddings</title>
@@ -8593,6 +8599,7 @@
       <url hash="290434e1">2021.emnlp-main.552</url>
       <bibkey>gao-etal-2021-simcse</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.552</doi>
+      <video href="2021.emnlp-main.552.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/SimCSE" additional="true">princeton-nlp/SimCSE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
@@ -8602,7 +8609,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/sts-benchmark">STS Benchmark</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semantic-textual-similarity-2012-2016">Semantic Textual Similarity (2012 - 2016)</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.emnlp-main.552.mp4"/>
     </paper>
     <paper id="553">
       <title>When is Wall a Pared and when a Muro?: Extracting Rules Governing Lexical Selection</title>
@@ -8615,8 +8621,8 @@
       <url hash="f56b9e90">2021.emnlp-main.553</url>
       <bibkey>chaudhary-etal-2021-wall</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.553</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
       <video href="2021.emnlp-main.553.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
     </paper>
     <paper id="554">
       <title>Aligning Actions Across Recipe Graphs</title>
@@ -8631,9 +8637,9 @@
       <url hash="257f187e">2021.emnlp-main.554</url>
       <bibkey>donatelli-etal-2021-aligning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.554</doi>
+      <video href="2021.emnlp-main.554.mp4"/>
       <pwccode url="https://github.com/coli-saar/ara" additional="false">coli-saar/ara</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/microsoft-research-multimodal-aligned-recipe">Microsoft Research Multimodal Aligned Recipe Corpus</pwcdataset>
-      <video href="2021.emnlp-main.554.mp4"/>
     </paper>
     <paper id="555">
       <title>Generating Datasets with Pretrained Language Models</title>
@@ -8644,12 +8650,12 @@
       <url hash="62dc576c">2021.emnlp-main.555</url>
       <bibkey>schick-schutze-2021-generating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.555</doi>
+      <video href="2021.emnlp-main.555.mp4"/>
       <pwccode url="https://github.com/timoschick/dino" additional="true">timoschick/dino</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sick">SICK</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sts-benchmark">STS Benchmark</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semantic-textual-similarity-2012-2016">Semantic Textual Similarity (2012 - 2016)</pwcdataset>
-      <video href="2021.emnlp-main.555.mp4"/>
     </paper>
     <paper id="556">
       <title>Continuous Entailment Patterns for Lexical Inference in Context</title>
@@ -8660,9 +8666,9 @@
       <url hash="c34c3f03">2021.emnlp-main.556</url>
       <bibkey>schmitt-schutze-2021-continuous</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.556</doi>
+      <video href="2021.emnlp-main.556.mp4"/>
       <pwccode url="https://github.com/mnschmit/conan" additional="false">mnschmit/conan</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sherliic">SherLIiC</pwcdataset>
-      <video href="2021.emnlp-main.556.mp4"/>
     </paper>
     <paper id="557">
       <title>Numeracy enhances the Literacy of Language Models</title>
@@ -8677,8 +8683,8 @@
       <revision id="1" href="2021.emnlp-main.557v1" hash="ecf6ae2a"/>
       <revision id="2" href="2021.emnlp-main.557v2" hash="a885570e" date="2021-11-09">Added acknowledgements.</revision>
       <doi>10.18653/v1/2021.emnlp-main.557</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/wikiconvert">WikiConvert</pwcdataset>
       <video href="2021.emnlp-main.557.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikiconvert">WikiConvert</pwcdataset>
     </paper>
     <paper id="558">
       <title>Students Who Study Together Learn Better: On the Importance of Collective Knowledge Distillation for Domain Transfer in Fact Verification</title>
@@ -8690,8 +8696,8 @@
       <url hash="60106c55">2021.emnlp-main.558</url>
       <bibkey>mithun-etal-2021-students</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.558</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/figer">FIGER</pwcdataset>
       <video href="2021.emnlp-main.558.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/figer">FIGER</pwcdataset>
     </paper>
     <paper id="559">
       <title><fixed-case>M</fixed-case>ulti<fixed-case>EURLEX</fixed-case> - A multi-lingual and multi-label legal document classification dataset for zero-shot cross-lingual transfer</title>
@@ -8705,10 +8711,10 @@
       <doi>10.18653/v1/2021.emnlp-main.559</doi>
       <revision id="1" href="2021.emnlp-main.559v1" hash="f03d0da8"/>
       <revision id="2" href="2021.emnlp-main.559v2" hash="293dd3d4" date="2022-04-29">Added missing acknowledgment.</revision>
+      <video href="2021.emnlp-main.559.mp4"/>
       <pwccode url="https://github.com/nlpaueb/multi-eurlex" additional="false">nlpaueb/multi-eurlex</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/eurlex57k">EURLEX57K</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
-      <video href="2021.emnlp-main.559.mp4"/>
     </paper>
     <paper id="560">
       <title>Joint Passage Ranking for Diverse Multi-Answer Retrieval</title>
@@ -8722,8 +8728,8 @@
       <url hash="b7938bc1">2021.emnlp-main.560</url>
       <bibkey>min-etal-2021-joint</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.560</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <video href="2021.emnlp-main.560.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
     </paper>
     <paper id="561">
       <title>Generative Context Pair Selection for Multi-hop Question Answering</title>
@@ -8739,9 +8745,9 @@
       <url hash="e38e9806">2021.emnlp-main.561</url>
       <bibkey>dua-etal-2021-generative</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.561</doi>
+      <video href="2021.emnlp-main.561.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihop">WikiHop</pwcdataset>
-      <video href="2021.emnlp-main.561.mp4"/>
     </paper>
     <paper id="562">
       <title>Synthetic Data Augmentation for Zero-Shot Cross-Lingual Question Answering</title>
@@ -8756,11 +8762,11 @@
       <url hash="f63f7199">2021.emnlp-main.562</url>
       <bibkey>riabi-etal-2021-synthetic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.562</doi>
+      <video href="2021.emnlp-main.562.mp4"/>
       <pwccode url="https://github.com/microsoft/unilm" additional="false">microsoft/unilm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.emnlp-main.562.mp4"/>
     </paper>
     <paper id="563">
       <title>Have You Seen That Number? Investigating Extrapolation in Question Answering Models</title>
@@ -8774,8 +8780,8 @@
       <url hash="64f481c2">2021.emnlp-main.563</url>
       <bibkey>kim-etal-2021-seen</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.563</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <video href="2021.emnlp-main.563.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
     </paper>
     <paper id="564">
       <title>Surface Form Competition: Why the Highest Probability Answer Isn’t Always Right</title>
@@ -8789,13 +8795,13 @@
       <url hash="10f8ae31">2021.emnlp-main.564</url>
       <bibkey>holtzman-etal-2021-surface</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.564</doi>
+      <video href="2021.emnlp-main.564.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/arc">ARC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/commonsenseqa">CommonsenseQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hellaswag">HellaSwag</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.564.mp4"/>
     </paper>
     <paper id="565">
       <title>Entity-Based Knowledge Conflicts in Question Answering</title>
@@ -8810,10 +8816,10 @@
       <url hash="18b7f5d7">2021.emnlp-main.565</url>
       <bibkey>longpre-etal-2021-entity</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.565</doi>
+      <video href="2021.emnlp-main.565.mp4"/>
       <pwccode url="https://github.com/apple/ml-knowledge-conflicts" additional="false">apple/ml-knowledge-conflicts</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsqa">NewsQA</pwcdataset>
-      <video href="2021.emnlp-main.565.mp4"/>
     </paper>
     <paper id="566">
       <title>Back-Training excels Self-Training at Unsupervised Domain Adaptation of Question Generation and Passage Retrieval</title>
@@ -8827,12 +8833,12 @@
       <attachment type="Software" hash="58cd5a39">2021.emnlp-main.566.Software.zip</attachment>
       <bibkey>kulshreshtha-etal-2021-back</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.566</doi>
+      <video href="2021.emnlp-main.566.mp4"/>
       <pwccode url="https://github.com/McGill-NLP/MLQuestions" additional="false">McGill-NLP/MLQuestions</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlquestions">MLQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/pubmedqa">PubMedQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vqg">VQG</pwcdataset>
-      <video href="2021.emnlp-main.566.mp4"/>
     </paper>
     <paper id="567">
       <title><fixed-case>DWUG</fixed-case>: A large Resource of Diachronic Word Usage Graphs in Four Languages</title>
@@ -8870,8 +8876,8 @@
       <url hash="62a1220b">2021.emnlp-main.569</url>
       <bibkey>wegmann-nguyen-2021-capture</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.569</doi>
-      <pwccode url="https://github.com/nlpsoc/stel" additional="false">nlpsoc/stel</pwccode>
       <video href="2021.emnlp-main.569.mp4"/>
+      <pwccode url="https://github.com/nlpsoc/stel" additional="false">nlpsoc/stel</pwccode>
     </paper>
     <paper id="570">
       <title>Evaluating the Morphosyntactic Well-formedness of Generated Texts</title>
@@ -8887,8 +8893,8 @@
       <url hash="aab8f609">2021.emnlp-main.570</url>
       <bibkey>pratapa-etal-2021-evaluating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.570</doi>
-      <pwccode url="https://github.com/adithya7/lambre" additional="false">adithya7/lambre</pwccode>
       <video href="2021.emnlp-main.570.mp4"/>
+      <pwccode url="https://github.com/adithya7/lambre" additional="false">adithya7/lambre</pwccode>
     </paper>
     <paper id="571">
       <title><fixed-case>AM</fixed-case>2i<fixed-case>C</fixed-case>o: Evaluating Word Meaning in Context across Low-Resource Languages with Adversarial Examples</title>
@@ -8902,12 +8908,12 @@
       <url hash="637ad1c2">2021.emnlp-main.571</url>
       <bibkey>liu-etal-2021-am2ico</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.571</doi>
+      <video href="2021.emnlp-main.571.mp4"/>
       <pwccode url="https://github.com/cambridgeltl/AM2iCo" additional="false">cambridgeltl/AM2iCo</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/am2ico">AM2iCo</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xl-wic">XL-WiC</pwcdataset>
-      <video href="2021.emnlp-main.571.mp4"/>
     </paper>
     <paper id="572">
       <title><fixed-case>C</fixed-case>ross<fixed-case>F</fixed-case>it: A Few-shot Learning Challenge for Cross-task Generalization in <fixed-case>NLP</fixed-case></title>
@@ -8920,13 +8926,13 @@
       <attachment type="Software" hash="058222f1">2021.emnlp-main.572.Software.rar</attachment>
       <bibkey>ye-etal-2021-crossfit</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.572</doi>
+      <video href="2021.emnlp-main.572.mp4"/>
       <pwccode url="https://github.com/INK-USC/CrossFit" additional="true">INK-USC/CrossFit</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/arc">ARC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quoref">Quoref</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
-      <video href="2021.emnlp-main.572.mp4"/>
     </paper>
     <paper id="573">
       <title>On the Influence of Masking Policies in Intermediate Pre-training</title>
@@ -8943,13 +8949,13 @@
       <url hash="f7e52f8b">2021.emnlp-main.573</url>
       <bibkey>ye-etal-2021-influence</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.573</doi>
+      <video href="2021.emnlp-main.573.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quartz">QuaRTz</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ropes">ROPES</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
-      <video href="2021.emnlp-main.573.mp4"/>
     </paper>
     <paper id="574">
       <title><fixed-case>V</fixed-case>al<fixed-case>N</fixed-case>orm Quantifies Semantics to Reveal Consistent Valence Biases Across Languages and Over Centuries</title>
@@ -8960,9 +8966,9 @@
       <url hash="e9e7c2b5">2021.emnlp-main.574</url>
       <bibkey>toney-caliskan-2021-valnorm</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.574</doi>
+      <video href="2021.emnlp-main.574.mp4"/>
       <pwccode url="https://github.com/autumntoney/ValNorm" additional="false">autumntoney/ValNorm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
-      <video href="2021.emnlp-main.574.mp4"/>
     </paper>
     <paper id="575">
       <title>Perturbation <fixed-case>C</fixed-case>heck<fixed-case>L</fixed-case>ists for Evaluating <fixed-case>NLG</fixed-case> Evaluation Metrics</title>
@@ -8976,8 +8982,8 @@
       <url hash="7bf7f3b6">2021.emnlp-main.575</url>
       <bibkey>sai-etal-2021-perturbation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.575</doi>
-      <pwccode url="https://github.com/iitmnlp/evaleval" additional="false">iitmnlp/evaleval</pwccode>
       <video href="2021.emnlp-main.575.mp4"/>
+      <pwccode url="https://github.com/iitmnlp/evaleval" additional="false">iitmnlp/evaleval</pwccode>
     </paper>
     <paper id="576">
       <title>Robust Open-Vocabulary Translation from Visual Text Representations</title>
@@ -8989,9 +8995,9 @@
       <url hash="2c2bd918">2021.emnlp-main.576</url>
       <bibkey>salesky-etal-2021-robust</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.576</doi>
+      <video href="2021.emnlp-main.576.mp4"/>
       <pwccode url="https://github.com/esalesky/visrep" additional="false">esalesky/visrep</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mtnt">MTNT</pwcdataset>
-      <video href="2021.emnlp-main.576.mp4"/>
     </paper>
     <paper id="577">
       <title>Don’t Go Far Off: An Empirical Study on Neural Poetry Translation</title>
@@ -9018,9 +9024,9 @@
       <url hash="1d4afce6">2021.emnlp-main.578</url>
       <bibkey>yang-etal-2021-improving-multilingual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.578</doi>
+      <video href="2021.emnlp-main.578.mp4"/>
       <pwccode url="https://github.com/yilinyang7/fairseq_multi_fix" additional="false">yilinyang7/fairseq_multi_fix</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/opus-100">OPUS-100</pwcdataset>
-      <video href="2021.emnlp-main.578.mp4"/>
     </paper>
     <paper id="579">
       <title>Learning Kernel-Smoothed Machine Translation with Retrieved Examples</title>
@@ -9035,9 +9041,9 @@
       <url hash="09faadd8">2021.emnlp-main.579</url>
       <bibkey>jiang-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.579</doi>
+      <video href="2021.emnlp-main.579.mp4"/>
       <pwccode url="https://github.com/jiangqn/kster" additional="false">jiangqn/kster</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
-      <video href="2021.emnlp-main.579.mp4"/>
     </paper>
     <paper id="580">
       <title>Uncertainty-Aware Balancing for Multilingual and Multi-Domain Neural Machine Translation Training</title>
@@ -9052,8 +9058,8 @@
       <url hash="38f0fc7a">2021.emnlp-main.580</url>
       <bibkey>wu-etal-2021-uncertainty</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.580</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
       <video href="2021.emnlp-main.580.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
     </paper>
     <paper id="581">
       <title>Universal Simultaneous Machine Translation with Mixture-of-Experts Wait-k Policy</title>
@@ -9064,8 +9070,8 @@
       <url hash="96ade5c8">2021.emnlp-main.581</url>
       <bibkey>zhang-feng-2021-universal</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.581</doi>
-      <pwccode url="https://github.com/ictnlp/moe-waitk" additional="false">ictnlp/moe-waitk</pwccode>
       <video href="2021.emnlp-main.581.mp4"/>
+      <pwccode url="https://github.com/ictnlp/moe-waitk" additional="false">ictnlp/moe-waitk</pwccode>
     </paper>
     <paper id="582">
       <title>How much coffee was consumed during <fixed-case>EMNLP</fixed-case> 2019? Fermi Problems: A New Reasoning Challenge for <fixed-case>AI</fixed-case></title>
@@ -9090,12 +9096,12 @@
       <url hash="e28c0f93">2021.emnlp-main.583</url>
       <bibkey>garg-moschitti-2021-will</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.583</doi>
+      <video href="2021.emnlp-main.583.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/asnq">ASNQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paralex">Paralex</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikiqa">WikiQA</pwcdataset>
-      <video href="2021.emnlp-main.583.mp4"/>
     </paper>
     <paper id="584">
       <title>Learning with Instance Bundles for Reading Comprehension</title>
@@ -9108,10 +9114,10 @@
       <url hash="2db9cc2c">2021.emnlp-main.584</url>
       <bibkey>dua-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.584</doi>
+      <video href="2021.emnlp-main.584.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quoref">Quoref</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ropes">ROPES</pwcdataset>
-      <video href="2021.emnlp-main.584.mp4"/>
     </paper>
     <paper id="585">
       <title>Explaining Answers with Entailment Trees</title>
@@ -9127,12 +9133,12 @@
       <url hash="52b7078a">2021.emnlp-main.585</url>
       <bibkey>dalvi-etal-2021-explaining</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.585</doi>
+      <video href="2021.emnlp-main.585.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/entailmentbank">EntailmentBank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/arc">ARC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/proofwriter">ProofWriter</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/worldtree">Worldtree</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/eqasc">eQASC</pwcdataset>
-      <video href="2021.emnlp-main.585.mp4"/>
     </paper>
     <paper id="586">
       <title><fixed-case>S</fixed-case>ituated<fixed-case>QA</fixed-case>: Incorporating Extra-Linguistic Contexts into <fixed-case>QA</fixed-case></title>
@@ -9143,10 +9149,10 @@
       <url hash="67f4a06c">2021.emnlp-main.586</url>
       <bibkey>zhang-choi-2021-situatedqa</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.586</doi>
+      <video href="2021.emnlp-main.586.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/situatedqa">SituatedQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.emnlp-main.586.mp4"/>
     </paper>
     <paper id="587">
       <title><fixed-case>C</fixed-case>onv<fixed-case>A</fixed-case>buse: Data, Analysis, and Benchmarks for Nuanced Abuse Detection in Conversational <fixed-case>AI</fixed-case></title>
@@ -9158,8 +9164,8 @@
       <url hash="b155b6d9">2021.emnlp-main.587</url>
       <bibkey>cercas-curry-etal-2021-convabuse</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.587</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/olid">OLID</pwcdataset>
       <video href="2021.emnlp-main.587.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/olid">OLID</pwcdataset>
     </paper>
     <paper id="588">
       <title>Conversational Multi-Hop Reasoning with Neural Commonsense Knowledge and Symbolic Logic Rules</title>
@@ -9173,8 +9179,8 @@
       <url hash="3fab6882">2021.emnlp-main.588</url>
       <bibkey>arabshahi-etal-2021-conversational</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.588</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <video href="2021.emnlp-main.588.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
     </paper>
     <paper id="589">
       <title>Towards Automatic Evaluation of Dialog Systems: A Model-Free Off-Policy Evaluation Approach</title>
@@ -9188,9 +9194,9 @@
       <url hash="038c959c">2021.emnlp-main.589</url>
       <bibkey>jiang-etal-2021-towards</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.589</doi>
+      <video href="2021.emnlp-main.589.mp4"/>
       <pwccode url="https://github.com/google-research/google-research" additional="false">google-research/google-research</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/convai2">ConvAI2</pwcdataset>
-      <video href="2021.emnlp-main.589.mp4"/>
     </paper>
     <paper id="590">
       <title>Continual Learning in Task-Oriented Dialogue Systems</title>
@@ -9209,9 +9215,9 @@
       <url hash="09fc2da8">2021.emnlp-main.590</url>
       <bibkey>madotto-etal-2021-continual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.590</doi>
+      <video href="2021.emnlp-main.590.mp4"/>
       <pwccode url="https://github.com/andreamad8/ToDCL" additional="false">andreamad8/ToDCL</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sgd">SGD</pwcdataset>
-      <video href="2021.emnlp-main.590.mp4"/>
     </paper>
     <paper id="591">
       <title>Multilingual and Cross-Lingual Intent Detection from Spoken Data</title>
@@ -9229,8 +9235,8 @@
       <url hash="055347ee">2021.emnlp-main.591</url>
       <bibkey>gerz-etal-2021-multilingual</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.591</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <video href="2021.emnlp-main.591.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
     </paper>
     <paper id="592">
       <title>Investigating Robustness of Dialog Models to Popular Figurative Language Constructs</title>
@@ -9243,10 +9249,10 @@
       <url hash="701271a2">2021.emnlp-main.592</url>
       <bibkey>jhamtani-etal-2021-investigating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.592</doi>
+      <video href="2021.emnlp-main.592.mp4"/>
       <pwccode url="https://github.com/vgtomahawk/dialog-fig-speech-robust" additional="false">vgtomahawk/dialog-fig-speech-robust</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.emnlp-main.592.mp4"/>
     </paper>
     <paper id="593">
       <title>Effective Sequence-to-Sequence Dialogue State Tracking</title>
@@ -9260,10 +9266,10 @@
       <url hash="a821d5ba">2021.emnlp-main.593</url>
       <bibkey>zhao-etal-2021-effective-sequence</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.593</doi>
+      <video href="2021.emnlp-main.593.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dialogue-state-tracking-challenge">Dialogue State Tracking Challenge</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-oz">Wizard-of-Oz</pwcdataset>
-      <video href="2021.emnlp-main.593.mp4"/>
     </paper>
     <paper id="594">
       <title><fixed-case>MS</fixed-case>ˆ2: Multi-Document Summarization of Medical Studies</title>
@@ -9298,8 +9304,8 @@
       <doi>10.18653/v1/2021.emnlp-main.595</doi>
       <revision id="1" href="2021.emnlp-main.595v1" hash="6d54687c"/>
       <revision id="2" href="2021.emnlp-main.595v2" hash="c820e9c2" date="2022-03-28">Fixed results in Table 7 for reproducability purposes.</revision>
-      <pwccode url="https://github.com/jmhessel/clipscore" additional="true">jmhessel/clipscore</pwccode>
       <video href="2021.emnlp-main.595.mp4"/>
+      <pwccode url="https://github.com/jmhessel/clipscore" additional="true">jmhessel/clipscore</pwccode>
     </paper>
     <paper id="596">
       <title>On the Challenges of Evaluating Compositional Explanations in Multi-Hop Inference: Relevance, Completeness, and Expert Ratings</title>
@@ -9312,9 +9318,9 @@
       <url hash="8adfa133">2021.emnlp-main.596</url>
       <bibkey>jansen-etal-2021-challenges</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.596</doi>
+      <video href="2021.emnlp-main.596.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/qasc">QASC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/worldtree">Worldtree</pwcdataset>
-      <video href="2021.emnlp-main.596.mp4"/>
     </paper>
     <paper id="597">
       <title><fixed-case>ESTER</fixed-case>: A Machine Reading Comprehension Dataset for Reasoning about Event Semantic Relations</title>
@@ -9330,8 +9336,8 @@
       <url hash="985aae2f">2021.emnlp-main.597</url>
       <bibkey>han-etal-2021-ester</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.597</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/torque">Torque</pwcdataset>
       <video href="2021.emnlp-main.597.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/torque">Torque</pwcdataset>
     </paper>
     <paper id="598">
       <title><fixed-case>RICA</fixed-case>: Evaluating Robust Inference Capabilities Based on Commonsense Axioms</title>
@@ -9362,10 +9368,10 @@
       <attachment type="Software" hash="c0285721">2021.emnlp-main.599.Software.zip</attachment>
       <bibkey>deng-etal-2021-compression</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.599</doi>
+      <video href="2021.emnlp-main.599.mp4"/>
       <pwccode url="https://github.com/tanyuqian/ctc-gen-eval" additional="false">tanyuqian/ctc-gen-eval</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.emnlp-main.599.mp4"/>
     </paper>
     <paper id="600">
       <title><fixed-case>MATE</fixed-case>: Multi-view Attention for Table Transformer Efficiency</title>
@@ -9378,11 +9384,11 @@
       <url hash="91232325">2021.emnlp-main.600</url>
       <bibkey>eisenschlos-etal-2021-mate</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.600</doi>
+      <video href="2021.emnlp-main.600.mp4"/>
       <pwccode url="https://github.com/google-research/tapas" additional="false">google-research/tapas</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hybridqa">HybridQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sqa">SQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tabfact">TabFact</pwcdataset>
-      <video href="2021.emnlp-main.600.mp4"/>
     </paper>
     <paper id="601">
       <title>Learning with Different Amounts of Annotation: From Zero to Many Labels</title>
@@ -9394,10 +9400,10 @@
       <url hash="d27b81bf">2021.emnlp-main.601</url>
       <bibkey>zhang-etal-2021-learning-different</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.601</doi>
+      <video href="2021.emnlp-main.601.mp4"/>
       <pwccode url="https://github.com/szhang42/uneven_training_data" additional="false">szhang42/uneven_training_data</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/chaosnli">ChaosNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.601.mp4"/>
     </paper>
     <paper id="602">
       <title>When Attention Meets Fast Recurrence: Training Language Models with Reduced Compute</title>
@@ -9407,10 +9413,11 @@
       <url hash="edcf9a70">2021.emnlp-main.602</url>
       <bibkey>lei-2021-attention</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.602</doi>
+      <video href="2021.emnlp-main.602.mp4"/>
       <pwccode url="https://github.com/asappresearch/sru" additional="false">asappresearch/sru</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/billion-word-benchmark">Billion Word Benchmark</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.emnlp-main.602.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="603">
       <title>Universal-<fixed-case>KD</fixed-case>: Attention-based Output-Grounded Intermediate Layer Knowledge Distillation</title>
@@ -9424,12 +9431,12 @@
       <url hash="2f8f7f56">2021.emnlp-main.603</url>
       <bibkey>wu-etal-2021-universal</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.603</doi>
+      <video href="2021.emnlp-main.603.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.603.mp4"/>
     </paper>
     <paper id="604">
       <title>Highly Parallel Autoregressive Entity Linking with Discriminative Correction</title>
@@ -9441,9 +9448,9 @@
       <url hash="0d81d37b">2021.emnlp-main.604</url>
       <bibkey>de-cao-etal-2021-highly</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.604</doi>
+      <video href="2021.emnlp-main.604.mp4"/>
       <pwccode url="https://github.com/nicola-decao/efficient-autoregressive-EL" additional="false">nicola-decao/efficient-autoregressive-EL</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/aida-conll-yago">AIDA CoNLL-YAGO</pwcdataset>
-      <video href="2021.emnlp-main.604.mp4"/>
     </paper>
     <paper id="605">
       <title>Word-Level Coreference Resolution</title>
@@ -9453,10 +9460,10 @@
       <url hash="37503770">2021.emnlp-main.605</url>
       <bibkey>dobrovolskii-2021-word</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.605</doi>
+      <video href="2021.emnlp-main.605.mp4"/>
       <pwccode url="https://github.com/vdobrovolskii/wl-coref" additional="false">vdobrovolskii/wl-coref</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2012-1">CoNLL-2012</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ontonotes-5-0">OntoNotes 5.0</pwcdataset>
-      <video href="2021.emnlp-main.605.mp4"/>
     </paper>
     <paper id="606">
       <title>A Secure and Efficient Federated Learning Framework for <fixed-case>NLP</fixed-case></title>
@@ -9487,9 +9494,9 @@
       <url hash="961adf51">2021.emnlp-main.607</url>
       <bibkey>pasupat-etal-2021-controllable</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.607</doi>
+      <video href="2021.emnlp-main.607.mp4"/>
       <pwccode url="https://github.com/google-research/language/tree/master/language/casper" additional="false">google-research/language</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mtop">MTOP</pwcdataset>
-      <video href="2021.emnlp-main.607.mp4"/>
     </paper>
     <paper id="608">
       <title>Constrained Language Models Yield Few-Shot Semantic Parsers</title>
@@ -9508,9 +9515,9 @@
       <url hash="de57de7a">2021.emnlp-main.608</url>
       <bibkey>shin-etal-2021-constrained</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.608</doi>
+      <video href="2021.emnlp-main.608.mp4"/>
       <pwccode url="https://github.com/microsoft/semantic_parsing_with_constrained_lm" additional="false">microsoft/semantic_parsing_with_constrained_lm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/break">BREAK</pwcdataset>
-      <video href="2021.emnlp-main.608.mp4"/>
     </paper>
     <paper id="609">
       <title><fixed-case>E</fixed-case>xpla<fixed-case>G</fixed-case>raphs: An Explanation Graph Generation Task for Structured Commonsense Reasoning</title>
@@ -9523,8 +9530,8 @@
       <url hash="6c027f9c">2021.emnlp-main.609</url>
       <bibkey>saha-etal-2021-explagraphs</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.609</doi>
-      <pwccode url="https://github.com/swarnaHub/ExplaGraphs" additional="false">swarnaHub/ExplaGraphs</pwccode>
       <video href="2021.emnlp-main.609.mp4"/>
+      <pwccode url="https://github.com/swarnaHub/ExplaGraphs" additional="false">swarnaHub/ExplaGraphs</pwccode>
     </paper>
     <paper id="610">
       <title>Connect-the-Dots: Bridging Semantics between Words and Definitions via Aligning Word Sense Inventories</title>
@@ -9539,10 +9546,10 @@
       <url hash="1817095b">2021.emnlp-main.610</url>
       <bibkey>yao-etal-2021-connect</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.610</doi>
+      <video href="2021.emnlp-main.610.mp4"/>
       <pwccode url="https://github.com/tencent-ailab/EMNLP21_SemEq" additional="true">tencent-ailab/EMNLP21_SemEq</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
-      <video href="2021.emnlp-main.610.mp4"/>
     </paper>
     <paper id="611">
       <title><fixed-case>LM</fixed-case>-Critic: Language Models for Unsupervised Grammatical Error Correction</title>
@@ -9554,12 +9561,12 @@
       <url hash="dd0f2b8a">2021.emnlp-main.611</url>
       <bibkey>yasunaga-etal-2021-lm</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.611</doi>
+      <video href="2021.emnlp-main.611.mp4"/>
       <pwccode url="https://github.com/michiyasunaga/LM-Critic" additional="true">michiyasunaga/LM-Critic</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2014-shared-task-grammatical-error">CoNLL-2014 Shared Task: Grammatical Error Correction</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gmeg-wiki">GMEG-wiki</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gmeg-yahoo">GMEG-yahoo</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/locness-corpus">WI-LOCNESS</pwcdataset>
-      <video href="2021.emnlp-main.611.mp4"/>
     </paper>
     <paper id="612">
       <title>Language-agnostic Representation from Multilingual Sentence Encoders for Cross-lingual Similarity Estimation</title>
@@ -9572,8 +9579,8 @@
       <url hash="2a67ff5d">2021.emnlp-main.612</url>
       <bibkey>tiyajamorn-etal-2021-language</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.612</doi>
-      <pwccode url="https://github.com/nattaptiy/qe_disentangled" additional="false">nattaptiy/qe_disentangled</pwccode>
       <video href="2021.emnlp-main.612.mp4"/>
+      <pwccode url="https://github.com/nattaptiy/qe_disentangled" additional="false">nattaptiy/qe_disentangled</pwccode>
     </paper>
     <paper id="613">
       <title>Classifying Dyads for Militarized Conflict Analysis</title>
@@ -9587,8 +9594,8 @@
       <url hash="c2008001">2021.emnlp-main.613</url>
       <bibkey>stoehr-etal-2021-classifying</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.613</doi>
-      <pwccode url="https://github.com/conflict-ai/conflictwiki" additional="false">conflict-ai/conflictwiki</pwccode>
       <video href="2021.emnlp-main.613.mp4"/>
+      <pwccode url="https://github.com/conflict-ai/conflictwiki" additional="false">conflict-ai/conflictwiki</pwccode>
     </paper>
     <paper id="614">
       <title>Point-of-Interest Type Prediction using Text and Images</title>
@@ -9599,9 +9606,9 @@
       <url hash="60365974">2021.emnlp-main.614</url>
       <bibkey>sanchez-villegas-aletras-2021-point</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.614</doi>
+      <video href="2021.emnlp-main.614.mp4"/>
       <pwccode url="https://github.com/danaesavi/poi-type-prediction" additional="false">danaesavi/poi-type-prediction</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
-      <video href="2021.emnlp-main.614.mp4"/>
     </paper>
     <paper id="615">
       <title>Come hither or go away? Recognising pre-electoral coalition signals in the news</title>
@@ -9631,8 +9638,8 @@
       <url hash="76dc4a17">2021.emnlp-main.616</url>
       <bibkey>zhang-etal-2021-howyoutagtweets</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.616</doi>
-      <pwccode url="https://github.com/polyusmart/personalized-hashtag-preferences" additional="false">polyusmart/personalized-hashtag-preferences</pwccode>
       <video href="2021.emnlp-main.616.mp4"/>
+      <pwccode url="https://github.com/polyusmart/personalized-hashtag-preferences" additional="false">polyusmart/personalized-hashtag-preferences</pwccode>
     </paper>
     <paper id="617">
       <title>Learning Neural Templates for Recommender Dialogue System</title>
@@ -9650,8 +9657,8 @@
       <url hash="47c035d8">2021.emnlp-main.617</url>
       <bibkey>liang-etal-2021-learning-neural</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.617</doi>
-      <pwccode url="https://github.com/jokieleung/ntrd" additional="false">jokieleung/ntrd</pwccode>
       <video href="2021.emnlp-main.617.mp4"/>
+      <pwccode url="https://github.com/jokieleung/ntrd" additional="false">jokieleung/ntrd</pwccode>
     </paper>
     <paper id="618">
       <title>Proxy Indicators for the Quality of Open-domain Dialogues</title>
@@ -9663,6 +9670,7 @@
       <url hash="abc74a8a">2021.emnlp-main.618</url>
       <bibkey>nedelchev-etal-2021-proxy</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.618</doi>
+      <video href="2021.emnlp-main.618.mp4"/>
       <pwccode url="https://github.com/smartdataanalytics/proxy_indicators" additional="false">smartdataanalytics/proxy_indicators</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
@@ -9674,7 +9682,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/usr-personachat">USR-PersonaChat</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/usr-topicalchat">USR-TopicalChat</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
-      <video href="2021.emnlp-main.618.mp4"/>
     </paper>
     <paper id="619">
       <title><tex-math>Q^{2}</tex-math>: <fixed-case>E</fixed-case>valuating Factual Consistency in Knowledge-Grounded Dialogues via Question Generation and Question Answering</title>
@@ -9689,10 +9696,10 @@
       <url hash="b7b3f546">2021.emnlp-main.619</url>
       <bibkey>honovich-etal-2021-q2</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.619</doi>
+      <video href="2021.emnlp-main.619.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.emnlp-main.619.mp4"/>
     </paper>
     <paper id="620">
       <title>Knowledge-Aware Graph-Enhanced <fixed-case>GPT</fixed-case>-2 for Dialogue State Tracking</title>
@@ -9704,9 +9711,9 @@
       <url hash="a3f4fee1">2021.emnlp-main.620</url>
       <bibkey>lin-etal-2021-knowledge</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.620</doi>
+      <video href="2021.emnlp-main.620.mp4"/>
       <pwccode url="https://github.com/linweizhedragon/knowledge-aware-graph-enhanced-gpt-2-for-dialogue-state-tracking" additional="false">linweizhedragon/knowledge-aware-graph-enhanced-gpt-2-for-dialogue-state-tracking</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
-      <video href="2021.emnlp-main.620.mp4"/>
     </paper>
     <paper id="621">
       <title>A Collaborative Multi-agent Reinforcement Learning Framework for Dialog Action Decomposition</title>
@@ -9717,8 +9724,8 @@
       <url hash="946e81be">2021.emnlp-main.621</url>
       <bibkey>wang-wong-2021-collaborative</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.621</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
       <video href="2021.emnlp-main.621.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
     </paper>
     <paper id="622">
       <title>Zero-Shot Dialogue State Tracking via Cross-Task Transfer</title>
@@ -9738,6 +9745,7 @@
       <url hash="6efc0fb4">2021.emnlp-main.622</url>
       <bibkey>lin-etal-2021-zero</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.622</doi>
+      <video href="2021.emnlp-main.622.mp4"/>
       <pwccode url="https://github.com/facebookresearch/Zero-Shot-DST" additional="false">facebookresearch/Zero-Shot-DST</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dream">DREAM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
@@ -9748,7 +9756,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/searchqa">SearchQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.emnlp-main.622.mp4"/>
     </paper>
     <paper id="623">
       <title>Uncertainty Measures in Neural Belief Tracking and the Effects on Dialogue Policy Performance</title>
@@ -9790,8 +9797,8 @@
       <attachment type="Software" hash="4829229c">2021.emnlp-main.625.Software.zip</attachment>
       <bibkey>liang-etal-2021-semantic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.625</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
       <video href="2021.emnlp-main.625.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
     </paper>
     <paper id="626">
       <title><fixed-case>AdapterDrop</fixed-case>: <fixed-case>O</fixed-case>n the Efficiency of Adapters in Transformers</title>
@@ -9807,8 +9814,8 @@
       <url hash="1a607544">2021.emnlp-main.626</url>
       <bibkey>ruckle-etal-2021-adapterdrop</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.626</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <video href="2021.emnlp-main.626.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
     </paper>
     <paper id="627">
       <title>Understanding and Overcoming the Challenges of Efficient Transformer Quantization</title>
@@ -9820,10 +9827,10 @@
       <url hash="55ffd085">2021.emnlp-main.627</url>
       <bibkey>bondarenko-etal-2021-understanding</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.627</doi>
+      <video href="2021.emnlp-main.627.mp4"/>
       <pwccode url="https://github.com/qualcomm-ai-research/transformer-quantization" additional="false">qualcomm-ai-research/transformer-quantization</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.emnlp-main.627.mp4"/>
     </paper>
     <paper id="628">
       <title><fixed-case>CAPE</fixed-case>: Context-Aware Private Embeddings for Private Language Learning</title>
@@ -9835,8 +9842,8 @@
       <url hash="78618ae7">2021.emnlp-main.628</url>
       <bibkey>plant-etal-2021-cape</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.628</doi>
-      <pwccode url="" additional="true"/>
       <video href="2021.emnlp-main.628.mp4"/>
+      <pwccode url="" additional="true"/>
     </paper>
     <paper id="629">
       <title>Text Detoxification using Large Pre-trained Neural Models</title>
@@ -9852,8 +9859,8 @@
       <url hash="b4578d6f">2021.emnlp-main.629</url>
       <bibkey>dale-etal-2021-text</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.629</doi>
-      <pwccode url="https://github.com/skoltech-nlp/detox" additional="false">skoltech-nlp/detox</pwccode>
       <video href="2021.emnlp-main.629.mp4"/>
+      <pwccode url="https://github.com/skoltech-nlp/detox" additional="false">skoltech-nlp/detox</pwccode>
     </paper>
     <paper id="630">
       <title>Document-Level Text Simplification: Dataset, Criteria and Baseline</title>
@@ -9865,10 +9872,10 @@
       <url hash="c72d2d5a">2021.emnlp-main.630</url>
       <bibkey>sun-etal-2021-document</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.630</doi>
+      <video href="2021.emnlp-main.630.mp4"/>
       <pwccode url="https://github.com/rlsnlp/document-level-text-simplification" additional="false">rlsnlp/document-level-text-simplification</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/newsela">Newsela</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikilarge">WikiLarge</pwcdataset>
-      <video href="2021.emnlp-main.630.mp4"/>
     </paper>
     <paper id="631">
       <title>A Bag of Tricks for Dialogue Summarization</title>
@@ -9880,10 +9887,10 @@
       <url hash="c8d26f1e">2021.emnlp-main.631</url>
       <bibkey>khalifa-etal-2021-bag</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.631</doi>
+      <video href="2021.emnlp-main.631.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/commongen">CommonGen</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.emnlp-main.631.mp4"/>
     </paper>
     <paper id="632">
       <title>Paraphrasing Compound Nominalizations</title>
@@ -9911,10 +9918,10 @@
       <url hash="53912d38">2021.emnlp-main.633</url>
       <bibkey>rebuffel-etal-2021-data</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.633</doi>
+      <video href="2021.emnlp-main.633.mp4"/>
       <pwccode url="https://github.com/ThomasScialom/QuestEval" additional="true">ThomasScialom/QuestEval</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikibio">WikiBio</pwcdataset>
-      <video href="2021.emnlp-main.633.mp4"/>
     </paper>
     <paper id="634">
       <title>Low-Rank Subspaces for Unsupervised Entity Linking</title>
@@ -9941,10 +9948,10 @@
       <url hash="566e0c62">2021.emnlp-main.635</url>
       <bibkey>li-etal-2021-tdeer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.635</doi>
+      <video href="2021.emnlp-main.635.mp4"/>
       <pwccode url="https://github.com/4ai/tdeer" additional="false">4ai/tdeer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/nyt11-hrl">NYT11-HRL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.emnlp-main.635.mp4"/>
     </paper>
     <paper id="636">
       <title>Extracting Event Temporal Relations via Hyperbolic Geometry</title>
@@ -9956,9 +9963,9 @@
       <url hash="298bb25e">2021.emnlp-main.636</url>
       <bibkey>tan-etal-2021-extracting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.636</doi>
+      <video href="2021.emnlp-main.636.mp4"/>
       <pwccode url="https://github.com/xingwei-warwick/hyper-event-temprel" additional="false">xingwei-warwick/hyper-event-temprel</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/tcr">TCR</pwcdataset>
-      <video href="2021.emnlp-main.636.mp4"/>
     </paper>
     <paper id="637">
       <title>Honey or Poison? Solving the Trigger Curse in Few-shot Event Detection via Causal Intervention</title>
@@ -9971,9 +9978,9 @@
       <url hash="c64bb571">2021.emnlp-main.637</url>
       <bibkey>chen-etal-2021-honey</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.637</doi>
+      <video href="2021.emnlp-main.637.mp4"/>
       <pwccode url="https://github.com/chen700564/causalfsed" additional="false">chen700564/causalfsed</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/maven">MAVEN</pwcdataset>
-      <video href="2021.emnlp-main.637.mp4"/>
     </paper>
     <paper id="638">
       <title>Back to the Basics: A Quantitative Analysis of Statistical and Graph-Based Term Weighting Schemes for Keyword Extraction</title>
@@ -9985,9 +9992,9 @@
       <url hash="5795ffdc">2021.emnlp-main.638</url>
       <bibkey>ushio-etal-2021-back</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.638</doi>
+      <video href="2021.emnlp-main.638.mp4"/>
       <pwccode url="https://github.com/asahi417/kex" additional="false">asahi417/kex</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/semeval-2018-task-9-hypernym-discovery">SemEval-2018 Task 9: Hypernym Discovery</pwcdataset>
-      <video href="2021.emnlp-main.638.mp4"/>
     </paper>
     <paper id="639">
       <title>Time-dependent Entity Embedding is not All You Need: A Re-evaluation of Temporal Knowledge Graph Completion Models under a Unified Framework</title>
@@ -10000,8 +10007,8 @@
       <url hash="2f6c17f9">2021.emnlp-main.639</url>
       <bibkey>han-etal-2021-time</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.639</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
       <video href="2021.emnlp-main.639.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
     </paper>
     <paper id="640">
       <title>Matching-oriented Embedding Quantization For Ad-hoc Retrieval</title>
@@ -10015,8 +10022,8 @@
       <url hash="cbffa923">2021.emnlp-main.640</url>
       <bibkey>xiao-etal-2021-matching</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.640</doi>
-      <pwccode url="https://github.com/microsoft/mopq" additional="false">microsoft/mopq</pwccode>
       <video href="2021.emnlp-main.640.mp4"/>
+      <pwccode url="https://github.com/microsoft/mopq" additional="false">microsoft/mopq</pwccode>
     </paper>
     <paper id="641">
       <title>Efficient Mind-Map Generation via Sequence-to-Graph and Reinforced Graph Refinement</title>
@@ -10044,10 +10051,10 @@
       <url hash="cd2e10c7">2021.emnlp-main.642</url>
       <bibkey>liu-etal-2021-deep</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.642</doi>
+      <video href="2021.emnlp-main.642.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.642.mp4"/>
     </paper>
     <paper id="643">
       <title>Balancing Methods for Multi-label Text Classification with Long-Tailed Class Distribution</title>
@@ -10061,9 +10068,9 @@
       <url hash="d8f1174f">2021.emnlp-main.643</url>
       <bibkey>huang-etal-2021-balancing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.643</doi>
+      <video href="2021.emnlp-main.643.mp4"/>
       <pwccode url="https://github.com/Roche/BalancedLossNLP" additional="true">Roche/BalancedLossNLP</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/reuters-21578">Reuters-21578</pwcdataset>
-      <video href="2021.emnlp-main.643.mp4"/>
     </paper>
     <paper id="644">
       <title><fixed-case>B</fixed-case>ayesian Topic Regression for Causal Inference</title>
@@ -10078,8 +10085,8 @@
       <revision id="1" href="2021.emnlp-main.644v1" hash="357e4b70"/>
       <revision id="2" href="2021.emnlp-main.644v2" hash="be683200" date="2021-11-29">Corrects a small typo in the text and in an unnumbered equation.</revision>
       <doi>10.18653/v1/2021.emnlp-main.644</doi>
-      <pwccode url="https://github.com/maximilianahrens/data" additional="false">maximilianahrens/data</pwccode>
       <video href="2021.emnlp-main.644.mp4"/>
+      <pwccode url="https://github.com/maximilianahrens/data" additional="false">maximilianahrens/data</pwccode>
     </paper>
     <paper id="645">
       <title>Enjoy the Salience: Towards Better Transformer-based Faithful Explanations with Word Salience</title>
@@ -10090,9 +10097,9 @@
       <url hash="f606d6c1">2021.emnlp-main.645</url>
       <bibkey>chrysostomou-aletras-2021-enjoy</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.645</doi>
+      <video href="2021.emnlp-main.645.mp4"/>
       <pwccode url="https://github.com/gchrysostomou/saloss" additional="false">gchrysostomou/saloss</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.645.mp4"/>
     </paper>
     <paper id="646">
       <title><fixed-case>W</fixed-case>hat’s in Your Head? <fixed-case>E</fixed-case>mergent Behaviour in Multi-Task Transformer Models</title>
@@ -10105,8 +10112,8 @@
       <url hash="d8f88eed">2021.emnlp-main.646</url>
       <bibkey>geva-etal-2021-whats</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.646</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <video href="2021.emnlp-main.646.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
     </paper>
     <paper id="647">
       <title>Don’t Search for a Search Method — Simple Heuristics Suffice for Adversarial Text Attacks</title>
@@ -10132,8 +10139,8 @@
       <url hash="96f19b38">2021.emnlp-main.648</url>
       <bibkey>bhardwaj-etal-2021-adversarial</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.648</doi>
-      <pwccode url="https://github.com/perubhardwaj/attributionattack" additional="false">perubhardwaj/attributionattack</pwccode>
       <video href="2021.emnlp-main.648.mp4"/>
+      <pwccode url="https://github.com/perubhardwaj/attributionattack" additional="false">perubhardwaj/attributionattack</pwccode>
     </paper>
     <paper id="649">
       <title>Locke’s Holiday: Belief Bias in Machine Reading</title>
@@ -10143,8 +10150,8 @@
       <url hash="0494c636">2021.emnlp-main.649</url>
       <bibkey>sogaard-2021-lockes</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.649</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <video href="2021.emnlp-main.649.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
     </paper>
     <paper id="650">
       <title>Sequence Length is a Domain: Length-based Overfitting in Transformer Models</title>
@@ -10169,9 +10176,9 @@
       <url hash="17cb1839">2021.emnlp-main.651</url>
       <bibkey>mozes-etal-2021-contrasting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.651</doi>
+      <video href="2021.emnlp-main.651.mp4"/>
       <pwccode url="https://github.com/maximilianmozes/human_adversaries" additional="false">maximilianmozes/human_adversaries</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.651.mp4"/>
     </paper>
     <paper id="652">
       <title>Is Information Density Uniform in Task-Oriented Dialogues?</title>
@@ -10184,9 +10191,9 @@
       <attachment type="Software" hash="de86f70c">2021.emnlp-main.652.Software.zip</attachment>
       <bibkey>giulianelli-etal-2021-information</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.652</doi>
+      <video href="2021.emnlp-main.652.mp4"/>
       <pwccode url="https://github.com/dmg-illc/uid-dialogue" additional="false">dmg-illc/uid-dialogue</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/photobook">PhotoBook</pwcdataset>
-      <video href="2021.emnlp-main.652.mp4"/>
     </paper>
     <paper id="653">
       <title>On Homophony and Rényi Entropy</title>
@@ -10199,8 +10206,8 @@
       <url hash="5ad1601b">2021.emnlp-main.653</url>
       <bibkey>pimentel-etal-2021-homophony</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.653</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/celex">CELEX</pwcdataset>
       <video href="2021.emnlp-main.653.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/celex">CELEX</pwcdataset>
     </paper>
     <paper id="654">
       <title>Synthetic Textual Features for the Large-Scale Detection of Basic-level Categories in <fixed-case>E</fixed-case>nglish and <fixed-case>M</fixed-case>andarin</title>
@@ -10226,9 +10233,9 @@
       <attachment type="Software" hash="cd55d0a1">2021.emnlp-main.655.Software.zip</attachment>
       <bibkey>sun-etal-2021-timetraveler</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.655</doi>
+      <video href="2021.emnlp-main.655.mp4"/>
       <pwccode url="https://github.com/jhl-hust/titer" additional="false">jhl-hust/titer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
-      <video href="2021.emnlp-main.655.mp4"/>
     </paper>
     <paper id="656">
       <title>Code-switched inspired losses for spoken dialog representations</title>
@@ -10253,9 +10260,9 @@
       <attachment type="Software" hash="27dc4f2d">2021.emnlp-main.657.Software.zip</attachment>
       <bibkey>guo-kok-2021-bique</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.657</doi>
+      <video href="2021.emnlp-main.657.mp4"/>
       <pwccode url="https://github.com/guojiapub/bique" additional="false">guojiapub/bique</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
-      <video href="2021.emnlp-main.657.mp4"/>
     </paper>
     <paper id="658">
       <title>Learning Neural Ordinary Equations for Forecasting Future Links on Temporal Knowledge Graphs</title>
@@ -10269,8 +10276,8 @@
       <url hash="0bcf4aaf">2021.emnlp-main.658</url>
       <bibkey>han-etal-2021-learning-neural</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.658</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
       <video href="2021.emnlp-main.658.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
     </paper>
     <paper id="659">
       <title><fixed-case>RAP</fixed-case>: <fixed-case>R</fixed-case>obustness-<fixed-case>A</fixed-case>ware <fixed-case>P</fixed-case>erturbations for Defending against Backdoor Attacks on <fixed-case>NLP</fixed-case> Models</title>
@@ -10284,9 +10291,9 @@
       <url hash="7cd6e2cd">2021.emnlp-main.659</url>
       <bibkey>yang-etal-2021-rap</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.659</doi>
+      <video href="2021.emnlp-main.659.mp4"/>
       <pwccode url="https://github.com/lancopku/rap" additional="false">lancopku/rap</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.659.mp4"/>
     </paper>
     <paper id="660">
       <title><fixed-case>FAME</fixed-case>: <fixed-case>F</fixed-case>eature-Based Adversarial Meta-Embeddings for Robust Input Representations</title>
@@ -10299,8 +10306,8 @@
       <url hash="87fcf6fe">2021.emnlp-main.660</url>
       <bibkey>lange-etal-2021-fame</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.660</doi>
-      <pwccode url="https://github.com/boschresearch/adversarial_meta_embeddings" additional="false">boschresearch/adversarial_meta_embeddings</pwccode>
       <video href="2021.emnlp-main.660.mp4"/>
+      <pwccode url="https://github.com/boschresearch/adversarial_meta_embeddings" additional="false">boschresearch/adversarial_meta_embeddings</pwccode>
     </paper>
     <paper id="661">
       <title>A Strong Baseline for Query Efficient Attacks in a Black Box Setting</title>
@@ -10312,9 +10319,9 @@
       <url hash="950f6912">2021.emnlp-main.661</url>
       <bibkey>maheshwary-etal-2021-strong</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.661</doi>
+      <video href="2021.emnlp-main.661.mp4"/>
       <pwccode url="https://github.com/rishabhmaheshwary/query-attack" additional="false">rishabhmaheshwary/query-attack</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.emnlp-main.661.mp4"/>
     </paper>
     <paper id="662">
       <title>Machine Translation Decoding beyond Beam Search</title>
@@ -10358,9 +10365,9 @@
       <url hash="2955ecdc">2021.emnlp-main.664</url>
       <bibkey>raganato-etal-2021-empirical</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.664</doi>
+      <video href="2021.emnlp-main.664.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/opus-100">OPUS-100</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2018">WMT 2018</pwcdataset>
-      <video href="2021.emnlp-main.664.mp4"/>
     </paper>
     <paper id="665">
       <title>Graph Algorithms for Multiparallel Word Alignment</title>
@@ -10375,8 +10382,8 @@
       <url hash="96387566">2021.emnlp-main.665</url>
       <bibkey>imanigooghari-etal-2021-graph</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.665</doi>
-      <pwccode url="https://github.com/cisnlp/graph-align" additional="false">cisnlp/graph-align</pwccode>
       <video href="2021.emnlp-main.665.mp4"/>
+      <pwccode url="https://github.com/cisnlp/graph-align" additional="false">cisnlp/graph-align</pwccode>
     </paper>
     <paper id="666">
       <title>Improving the Quality Trade-Off for Neural Machine Translation Multi-Domain Adaptation</title>
@@ -10414,8 +10421,8 @@
       <url hash="ee99e4db">2021.emnlp-main.668</url>
       <bibkey>yu-joty-2021-effective</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.668</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <video href="2021.emnlp-main.668.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
     </paper>
     <paper id="669">
       <title>Rethinking Data Augmentation for Low-Resource Neural Machine Translation: A Multi-Task Learning Approach</title>
@@ -10428,8 +10435,8 @@
       <url hash="dc420392">2021.emnlp-main.669</url>
       <bibkey>sanchez-cartagena-etal-2021-rethinking</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.669</doi>
-      <pwccode url="https://github.com/transducens/mtl-da-emnlp" additional="false">transducens/mtl-da-emnlp</pwccode>
       <video href="2021.emnlp-main.669.mp4"/>
+      <pwccode url="https://github.com/transducens/mtl-da-emnlp" additional="false">transducens/mtl-da-emnlp</pwccode>
     </paper>
     <paper id="670">
       <title>Wino-<fixed-case>X</fixed-case>: Multilingual <fixed-case>W</fixed-case>inograd Schemas for Commonsense Reasoning and Coreference Resolution</title>
@@ -10441,9 +10448,9 @@
       <attachment type="Software" hash="837e5225">2021.emnlp-main.670.Software.zip</attachment>
       <bibkey>emelin-sennrich-2021-wino</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.670</doi>
+      <video href="2021.emnlp-main.670.mp4"/>
       <pwccode url="https://github.com/demelin/wino-x" additional="false">demelin/wino-x</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/winogrande">WinoGrande</pwcdataset>
-      <video href="2021.emnlp-main.670.mp4"/>
     </paper>
     <paper id="671">
       <title>One Source, Two Targets: <fixed-case>C</fixed-case>hallenges and Rewards of Dual Decoding</title>
@@ -10454,8 +10461,8 @@
       <url hash="20e34fa8">2021.emnlp-main.671</url>
       <bibkey>xu-yvon-2021-one</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.671</doi>
-      <pwccode url="https://github.com/jitao-xu/dual-decoding" additional="false">jitao-xu/dual-decoding</pwccode>
       <video href="2021.emnlp-main.671.mp4"/>
+      <pwccode url="https://github.com/jitao-xu/dual-decoding" additional="false">jitao-xu/dual-decoding</pwccode>
     </paper>
     <paper id="672">
       <title>Discrete and Soft Prompting for Multilingual Models</title>
@@ -10466,8 +10473,8 @@
       <url hash="ab628c18">2021.emnlp-main.672</url>
       <bibkey>zhao-schutze-2021-discrete</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.672</doi>
-      <pwccode url="https://github.com/mprompting/xlmrprompt" additional="false">mprompting/xlmrprompt</pwccode>
       <video href="2021.emnlp-main.672.mp4"/>
+      <pwccode url="https://github.com/mprompting/xlmrprompt" additional="false">mprompting/xlmrprompt</pwccode>
     </paper>
     <paper id="673">
       <title>Vision Matters When It Should: Sanity Checking Multimodal Machine Translation Models</title>
@@ -10479,10 +10486,10 @@
       <url hash="5e6e1c92">2021.emnlp-main.673</url>
       <bibkey>li-etal-2021-vision</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.673</doi>
+      <video href="2021.emnlp-main.673.mp4"/>
       <pwccode url="https://github.com/jiaodali/vision-matters-when-it-should" additional="false">jiaodali/vision-matters-when-it-should</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multisense">MultiSense</pwcdataset>
-      <video href="2021.emnlp-main.673.mp4"/>
     </paper>
     <paper id="674">
       <title>Efficient Inference for Multilingual Neural Machine Translation</title>
@@ -10496,8 +10503,8 @@
       <url hash="9f3a7029">2021.emnlp-main.674</url>
       <bibkey>berard-etal-2021-efficient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.674</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/paracrawl">ParaCrawl</pwcdataset>
       <video href="2021.emnlp-main.674.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/paracrawl">ParaCrawl</pwcdataset>
     </paper>
     <paper id="675">
       <title>Role of <fixed-case>L</fixed-case>anguage <fixed-case>R</fixed-case>elatedness in <fixed-case>M</fixed-case>ultilingual <fixed-case>F</fixed-case>ine-tuning of <fixed-case>L</fixed-case>anguage <fixed-case>M</fixed-case>odels: <fixed-case>A</fixed-case> <fixed-case>C</fixed-case>ase <fixed-case>S</fixed-case>tudy in <fixed-case>I</fixed-case>ndo-<fixed-case>A</fixed-case>ryan <fixed-case>L</fixed-case>anguages</title>
@@ -10512,9 +10519,9 @@
       <attachment type="Software" hash="ab82f874">2021.emnlp-main.675.Software.zip</attachment>
       <bibkey>dhamecha-etal-2021-role</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.675</doi>
+      <video href="2021.emnlp-main.675.mp4"/>
       <pwccode url="https://github.com/ibm/indo-aryan-language-family-model" additional="false">ibm/indo-aryan-language-family-model</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/indicglue">IndicGLUE</pwcdataset>
-      <video href="2021.emnlp-main.675.mp4"/>
     </paper>
     <paper id="676">
       <title>Comparing Feature-Engineering and Feature-Learning Approaches for Multilingual Translationese Classification</title>
@@ -10539,9 +10546,9 @@
       <url hash="4f89bfde">2021.emnlp-main.677</url>
       <bibkey>provilkov-malinin-2021-multi</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.677</doi>
+      <video href="2021.emnlp-main.677.mp4"/>
       <pwccode url="https://github.com/yandex-research/msr" additional="false">yandex-research/msr</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
-      <video href="2021.emnlp-main.677.mp4"/>
     </paper>
     <paper id="678">
       <title>Cross-Policy Compliance Detection via Question Answering</title>
@@ -10553,9 +10560,9 @@
       <url hash="e4acd2d5">2021.emnlp-main.678</url>
       <bibkey>saeidi-etal-2021-cross</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.678</doi>
+      <video href="2021.emnlp-main.678.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sharc">ShARC</pwcdataset>
-      <video href="2021.emnlp-main.678.mp4"/>
     </paper>
     <paper id="679">
       <title>Meta-<fixed-case>LMTC</fixed-case>: Meta-Learning for Large-Scale Multi-Label Text Classification</title>
@@ -10570,8 +10577,8 @@
       <url hash="7e5317de">2021.emnlp-main.679</url>
       <bibkey>wang-etal-2021-meta-lmtc</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.679</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/eurlex57k">EURLEX57K</pwcdataset>
       <video href="2021.emnlp-main.679.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/eurlex57k">EURLEX57K</pwcdataset>
     </paper>
     <paper id="680">
       <title>Unsupervised Multi-View Post-<fixed-case>OCR</fixed-case> Error Correction With Language Models</title>
@@ -10595,8 +10602,8 @@
       <url hash="2d960f40">2021.emnlp-main.681</url>
       <bibkey>he-2021-parallel</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.681</doi>
-      <pwccode url="https://github.com/nlpcode/cbart" additional="false">nlpcode/cbart</pwccode>
       <video href="2021.emnlp-main.681.mp4"/>
+      <pwccode url="https://github.com/nlpcode/cbart" additional="false">nlpcode/cbart</pwccode>
     </paper>
     <paper id="682">
       <title><fixed-case>BERT</fixed-case>-Beta: A Proactive Probabilistic Approach to Text Moderation</title>
@@ -10622,8 +10629,8 @@
       <url hash="2dc5b8b2">2021.emnlp-main.683</url>
       <bibkey>ghosal-etal-2021-stack</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.683</doi>
-      <pwccode url="https://github.com/declare-lab/sentence-ordering" additional="false">declare-lab/sentence-ordering</pwccode>
       <video href="2021.emnlp-main.683.mp4"/>
+      <pwccode url="https://github.com/declare-lab/sentence-ordering" additional="false">declare-lab/sentence-ordering</pwccode>
     </paper>
     <paper id="684">
       <title>Preventing Author Profiling through Zero-Shot Multilingual Back-Translation</title>
@@ -10638,9 +10645,9 @@
       <url hash="bf2b4076">2021.emnlp-main.684</url>
       <bibkey>adelani-etal-2021-preventing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.684</doi>
+      <video href="2021.emnlp-main.684.mp4"/>
       <pwccode url="https://github.com/uds-lsv/author-profiling-prevention-bt" additional="false">uds-lsv/author-profiling-prevention-bt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
-      <video href="2021.emnlp-main.684.mp4"/>
     </paper>
     <paper id="685">
       <title><fixed-case>C</fixed-case>ode<fixed-case>T</fixed-case>5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and Generation</title>
@@ -10653,11 +10660,11 @@
       <url hash="56481874">2021.emnlp-main.685</url>
       <bibkey>wang-etal-2021-codet5</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.685</doi>
+      <video href="2021.emnlp-main.685.mp4"/>
       <pwccode url="https://github.com/salesforce/codet5" additional="false">salesforce/codet5</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/concode">CONCODE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/codesearchnet">CodeSearchNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/codexglue">CodeXGLUE</pwcdataset>
-      <video href="2021.emnlp-main.685.mp4"/>
     </paper>
     <paper id="686">
       <title>Detect and Classify – Joint Span Detection and Classification for Health Outcomes</title>
@@ -10671,8 +10678,8 @@
       <attachment type="Software" hash="ff027604">2021.emnlp-main.686.Software.zip</attachment>
       <bibkey>abaho-etal-2021-detect</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.686</doi>
-      <pwccode url="https://github.com/MichealAbaho/Label-Context-Aware-Attention-Model" additional="false">MichealAbaho/Label-Context-Aware-Attention-Model</pwccode>
       <video href="2021.emnlp-main.686.mp4"/>
+      <pwccode url="https://github.com/MichealAbaho/Label-Context-Aware-Attention-Model" additional="false">MichealAbaho/Label-Context-Aware-Attention-Model</pwccode>
     </paper>
     <paper id="687">
       <title><fixed-case>M</fixed-case>ulti-Class Grammatical Error Detection for Correction: <fixed-case>A</fixed-case> Tale of Two Systems</title>
@@ -10685,8 +10692,8 @@
       <url hash="5d98ddfc">2021.emnlp-main.687</url>
       <bibkey>yuan-etal-2021-multi</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.687</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/fce">FCE</pwcdataset>
       <video href="2021.emnlp-main.687.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fce">FCE</pwcdataset>
     </paper>
     <paper id="688">
       <title>Towards Zero-shot Commonsense Reasoning with Self-supervised Refinement of Language Models</title>
@@ -10697,12 +10704,12 @@
       <url hash="0f483603">2021.emnlp-main.688</url>
       <bibkey>klein-nabi-2021-towards</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.688</doi>
+      <video href="2021.emnlp-main.688.mp4"/>
       <pwccode url="https://github.com/sap-samples/emnlp2021-contrastive-refinement" additional="false">sap-samples/emnlp2021-contrastive-refinement</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gap-coreference-dataset">GAP Coreference Dataset</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winobias">WinoBias</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winogrande">WinoGrande</pwcdataset>
-      <video href="2021.emnlp-main.688.mp4"/>
     </paper>
     <paper id="689">
       <title>To Share or not to Share: <fixed-case>P</fixed-case>redicting Sets of Sources for Model Transfer Learning</title>
@@ -10715,8 +10722,8 @@
       <url hash="787d85a3">2021.emnlp-main.689</url>
       <bibkey>lange-etal-2021-share</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.689</doi>
-      <pwccode url="https://github.com/boschresearch/predicting_sets_of_sources" additional="false">boschresearch/predicting_sets_of_sources</pwccode>
       <video href="2021.emnlp-main.689.mp4"/>
+      <pwccode url="https://github.com/boschresearch/predicting_sets_of_sources" additional="false">boschresearch/predicting_sets_of_sources</pwccode>
     </paper>
     <paper id="690">
       <title>Self-Supervised Detection of Contextual Synonyms in a Multi-Class Setting: Phenotype Annotation Use Case</title>
@@ -10761,8 +10768,8 @@
       <url hash="c18020e9">2021.emnlp-main.692</url>
       <bibkey>hamalainen-etal-2021-finnish</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.692</doi>
-      <pwccode url="https://github.com/rootroo-ltd/finnishdialectidentification" additional="false">rootroo-ltd/finnishdialectidentification</pwccode>
       <video href="2021.emnlp-main.692.mp4"/>
+      <pwccode url="https://github.com/rootroo-ltd/finnishdialectidentification" additional="false">rootroo-ltd/finnishdialectidentification</pwccode>
     </paper>
     <paper id="693">
       <title><fixed-case>E</fixed-case>nglish Machine Reading Comprehension Datasets: A Survey</title>
@@ -10774,6 +10781,7 @@
       <url hash="d3058e6e">2021.emnlp-main.693</url>
       <bibkey>dzendzik-etal-2021-english</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.693</doi>
+      <video href="2021.emnlp-main.693.mp4"/>
       <pwccode url="https://github.com/dariad/rczoo" additional="false">dariad/rczoo</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/arc">ARC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/amazonqa">AmazonQA</pwcdataset>
@@ -10821,7 +10829,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/wikireading">WikiReading</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/worldtree">Worldtree</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/emrqa">emrQA</pwcdataset>
-      <video href="2021.emnlp-main.693.mp4"/>
     </paper>
     <paper id="694">
       <title>Expanding End-to-End Question Answering on Differentiable Knowledge Graphs with Intersection</title>
@@ -10833,9 +10840,9 @@
       <url hash="3c5f48cd">2021.emnlp-main.694</url>
       <bibkey>sen-etal-2021-expanding</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.694</doi>
+      <video href="2021.emnlp-main.694.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/complexwebquestions">ComplexWebQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestionssp">WebQuestionsSP</pwcdataset>
-      <video href="2021.emnlp-main.694.mp4"/>
     </paper>
     <paper id="695">
       <title>Structured Context and High-Coverage Grammar for Conversational Question Answering over Knowledge Graphs</title>
@@ -10847,9 +10854,9 @@
       <url hash="d5c1d6ee">2021.emnlp-main.695</url>
       <bibkey>marion-etal-2021-structured</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.695</doi>
+      <video href="2021.emnlp-main.695.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/csqa">CSQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/convquestions">ConvQuestions</pwcdataset>
-      <video href="2021.emnlp-main.695.mp4"/>
     </paper>
     <paper id="696">
       <title>Improving Question Answering Model Robustness with Synthetic Adversarial Data Generation</title>
@@ -10864,11 +10871,11 @@
       <url hash="3ec9154d">2021.emnlp-main.696</url>
       <bibkey>bartolo-etal-2021-improving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.696</doi>
+      <video href="2021.emnlp-main.696.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/adversarialqa">AdversarialQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrqa-2019">MRQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.emnlp-main.696.mp4"/>
     </paper>
     <paper id="697">
       <title><fixed-case>B</fixed-case>elief<fixed-case>B</fixed-case>ank: Adding Memory to a Pre-Trained Language Model for a Systematic Notion of Belief</title>
@@ -10881,8 +10888,8 @@
       <url hash="5ab09962">2021.emnlp-main.697</url>
       <bibkey>kassner-etal-2021-beliefbank</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.697</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <video href="2021.emnlp-main.697.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
     </paper>
     <paper id="698">
       <title><fixed-case>MLEC-QA</fixed-case>: <fixed-case>A</fixed-case> <fixed-case>C</fixed-case>hinese <fixed-case>M</fixed-case>ulti-<fixed-case>C</fixed-case>hoice <fixed-case>B</fixed-case>iomedical <fixed-case>Q</fixed-case>uestion <fixed-case>A</fixed-case>nswering <fixed-case>D</fixed-case>ataset</title>
@@ -10894,9 +10901,9 @@
       <url hash="2a6e4581">2021.emnlp-main.698</url>
       <bibkey>li-etal-2021-mlec</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.698</doi>
+      <video href="2021.emnlp-main.698.mp4"/>
       <pwccode url="https://github.com/judenpech/mlec-qa" additional="false">judenpech/mlec-qa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/headqa">HeadQA</pwcdataset>
-      <video href="2021.emnlp-main.698.mp4"/>
     </paper>
     <paper id="699">
       <title><fixed-case>I</fixed-case>ndo<fixed-case>NLG</fixed-case>: Benchmark and Resources for Evaluating <fixed-case>I</fixed-case>ndonesian Natural Language Generation</title>
@@ -10917,12 +10924,12 @@
       <url hash="5f98a1d4">2021.emnlp-main.699</url>
       <bibkey>cahyawijaya-etal-2021-indonlg</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.699</doi>
+      <video href="2021.emnlp-main.699.mp4"/>
       <pwccode url="" additional="true"/>
       <pwcdataset url="https://paperswithcode.com/dataset/indonlg">IndoNLG</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gem">GEM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/liputan6">Liputan6</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
-      <video href="2021.emnlp-main.699.mp4"/>
     </paper>
     <paper id="700">
       <title>Is Multi-Hop Reasoning Really Explainable? Towards Benchmarking Reasoning Interpretability</title>
@@ -10938,8 +10945,8 @@
       <url hash="4e12918c">2021.emnlp-main.700</url>
       <bibkey>lv-etal-2021-multi</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.700</doi>
-      <pwccode url="https://github.com/THU-KEG/BIMR" additional="false">THU-KEG/BIMR</pwccode>
       <video href="2021.emnlp-main.700.mp4"/>
+      <pwccode url="https://github.com/THU-KEG/BIMR" additional="false">THU-KEG/BIMR</pwccode>
     </paper>
     <paper id="701">
       <title>Global Explainability of <fixed-case>BERT</fixed-case>-Based Evaluation Metrics by Disentangling along Linguistic Factors</title>
@@ -10951,9 +10958,9 @@
       <url hash="bf414303">2021.emnlp-main.701</url>
       <bibkey>kaster-etal-2021-global</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.701</doi>
+      <video href="2021.emnlp-main.701.mp4"/>
       <pwccode url="https://github.com/steffeneger/global-explainability-metrics" additional="false">steffeneger/global-explainability-metrics</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/paws">PAWS</pwcdataset>
-      <video href="2021.emnlp-main.701.mp4"/>
     </paper>
     <paper id="702">
       <title>Exploring Underexplored Limitations of Cross-Domain Text-to-<fixed-case>SQL</fixed-case> Generalization</title>
@@ -10965,8 +10972,8 @@
       <url hash="4b6052c0">2021.emnlp-main.702</url>
       <bibkey>gan-etal-2021-exploring</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.702</doi>
-      <pwccode url="https://github.com/ygan/spider-dk" additional="false">ygan/spider-dk</pwccode>
       <video href="2021.emnlp-main.702.mp4"/>
+      <pwccode url="https://github.com/ygan/spider-dk" additional="false">ygan/spider-dk</pwccode>
     </paper>
     <paper id="703">
       <title>What happens if you treat ordinal ratings as interval data? Human evaluations in <fixed-case>NLP</fixed-case> are even more under-powered than you think</title>
@@ -10989,9 +10996,9 @@
       <url hash="f7e486db">2021.emnlp-main.704</url>
       <bibkey>vanmassenhove-etal-2021-neutral</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.704</doi>
+      <video href="2021.emnlp-main.704.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winobias">WinoBias</pwcdataset>
-      <video href="2021.emnlp-main.704.mp4"/>
     </paper>
     <paper id="705">
       <title>Benchmarking Commonsense Knowledge Base Population with an Effective Evaluation Dataset</title>
@@ -11007,10 +11014,10 @@
       <url hash="4c7070dc">2021.emnlp-main.705</url>
       <bibkey>fang-etal-2021-benchmarking</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.705</doi>
-      <pwccode url="https://github.com/hkust-knowcomp/cskb-population" additional="false">hkust-knowcomp/cskb-population</pwccode>
+      <video href="2021.emnlp-main.705.mp4"/>
+      <pwccode url="https://github.com/hkust-knowcomp/cskb-population" additional="true">hkust-knowcomp/cskb-population</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glucose">GLUCOSE</pwcdataset>
-      <video href="2021.emnlp-main.705.mp4"/>
     </paper>
     <paper id="706">
       <title>Enhancing the Context Representation in Similarity-based Word Sense Disambiguation</title>
@@ -11023,8 +11030,8 @@
       <attachment type="Software" hash="71b95bd0">2021.emnlp-main.706.Software.zip</attachment>
       <bibkey>wang-etal-2021-enhancing-context</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.706</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
       <video href="2021.emnlp-main.706.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
     </paper>
     <paper id="707">
       <title>Data Augmentation with Hierarchical <fixed-case>SQL</fixed-case>-to-Question Generation for Cross-domain Text-to-<fixed-case>SQL</fixed-case> Parsing</title>
@@ -11041,10 +11048,10 @@
       <url hash="cd38f313">2021.emnlp-main.707</url>
       <bibkey>wu-etal-2021-data</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.707</doi>
+      <video href="2021.emnlp-main.707.mp4"/>
       <pwccode url="https://github.com/PaddlePaddle/Research" additional="false">PaddlePaddle/Research</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
-      <video href="2021.emnlp-main.707.mp4"/>
     </paper>
     <paper id="708">
       <title><fixed-case>SPARQL</fixed-case>ing Database Queries from Intermediate Question Decompositions</title>
@@ -11056,10 +11063,10 @@
       <attachment type="Software" hash="d580fc77">2021.emnlp-main.708.Software.zip</attachment>
       <bibkey>saparina-osokin-2021-sparqling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.708</doi>
+      <video href="2021.emnlp-main.708.mp4"/>
       <pwccode url="https://github.com/yandex-research/sparqling-queries" additional="false">yandex-research/sparqling-queries</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/break">BREAK</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
-      <video href="2021.emnlp-main.708.mp4"/>
     </paper>
     <paper id="709">
       <title>Time-aware Graph Neural Network for Entity Alignment between Temporal Knowledge Graphs</title>
@@ -11071,9 +11078,9 @@
       <url hash="e9ac0f22">2021.emnlp-main.709</url>
       <bibkey>xu-etal-2021-time</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.709</doi>
+      <video href="2021.emnlp-main.709.mp4"/>
       <pwccode url="https://github.com/soledad921/tea-gnn" additional="false">soledad921/tea-gnn</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/yago">YAGO</pwcdataset>
-      <video href="2021.emnlp-main.709.mp4"/>
     </paper>
     <paper id="710">
       <title>Cross-Domain Label-Adaptive Stance Detection</title>
@@ -11086,8 +11093,8 @@
       <url hash="0a04bf62">2021.emnlp-main.710</url>
       <bibkey>hardalov-etal-2021-cross</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.710</doi>
-      <pwccode url="https://github.com/checkstep/mole-stance" additional="false">checkstep/mole-stance</pwccode>
       <video href="2021.emnlp-main.710.mp4"/>
+      <pwccode url="https://github.com/checkstep/mole-stance" additional="false">checkstep/mole-stance</pwccode>
     </paper>
     <paper id="711">
       <title>Text <fixed-case>A</fixed-case>uto<fixed-case>A</fixed-case>ugment: Learning Compositional Augmentation Policy for Text Classification</title>
@@ -11101,10 +11108,10 @@
       <url hash="1fddc3bb">2021.emnlp-main.711</url>
       <bibkey>ren-etal-2021-text</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.711</doi>
+      <video href="2021.emnlp-main.711.mp4"/>
       <pwccode url="https://github.com/lancopku/text-autoaugment" additional="false">lancopku/text-autoaugment</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.711.mp4"/>
     </paper>
     <paper id="712">
       <title>Distilling Relation Embeddings from Pretrained Language Models</title>
@@ -11116,10 +11123,10 @@
       <url hash="c43be5b6">2021.emnlp-main.712</url>
       <bibkey>ushio-etal-2021-distilling</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.712</doi>
+      <video href="2021.emnlp-main.712.mp4"/>
       <pwccode url="https://github.com/asahi417/relbert" additional="false">asahi417/relbert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/evalution">EVALution</pwcdataset>
-      <video href="2021.emnlp-main.712.mp4"/>
     </paper>
     <paper id="713">
       <title>Avoiding Inference Heuristics in Few-shot Prompt-based Finetuning</title>
@@ -11132,11 +11139,11 @@
       <url hash="87d8a3f1">2021.emnlp-main.713</url>
       <bibkey>utama-etal-2021-avoiding</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.713</doi>
+      <video href="2021.emnlp-main.713.mp4"/>
       <pwccode url="https://github.com/ukplab/emnlp2021-prompt-ft-heuristics" additional="false">ukplab/emnlp2021-prompt-ft-heuristics</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws">PAWS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.713.mp4"/>
     </paper>
     <paper id="714">
       <title>A Differentiable Relaxation of Graph Segmentation and Alignment for <fixed-case>AMR</fixed-case> Parsing</title>
@@ -11148,8 +11155,8 @@
       <url hash="3c4f9833">2021.emnlp-main.714</url>
       <bibkey>lyu-etal-2021-differentiable</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.714</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/ldc2017t10">LDC2017T10</pwcdataset>
       <video href="2021.emnlp-main.714.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ldc2017t10">LDC2017T10</pwcdataset>
     </paper>
     <paper id="715">
       <title>Integrating Personalized <fixed-case>P</fixed-case>age<fixed-case>R</fixed-case>ank into Neural Word Sense Disambiguation</title>
@@ -11161,9 +11168,9 @@
       <url hash="4e2100ea">2021.emnlp-main.715</url>
       <bibkey>el-sheikh-etal-2021-integrating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.715</doi>
+      <video href="2021.emnlp-main.715.mp4"/>
       <pwccode url="https://github.com/sapienzanlp/neural-pagerank-wsd" additional="false">sapienzanlp/neural-pagerank-wsd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
-      <video href="2021.emnlp-main.715.mp4"/>
     </paper>
     <paper id="716">
       <title>Cross-lingual Sentence Embedding using Multi-Task Learning</title>
@@ -11177,11 +11184,11 @@
       <url hash="3d526338">2021.emnlp-main.716</url>
       <bibkey>goswami-etal-2021-cross</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.716</doi>
+      <video href="2021.emnlp-main.716.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bucc">BUCC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.emnlp-main.716.mp4"/>
     </paper>
     <paper id="717">
       <title><fixed-case>NB</fixed-case>-<fixed-case>MLM</fixed-case>: Efficient Domain Adaptation of Masked Language Models for Sentiment Analysis</title>
@@ -11193,8 +11200,8 @@
       <url hash="e329a0a1">2021.emnlp-main.717</url>
       <bibkey>arefyev-etal-2021-nb</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.717</doi>
-      <pwccode url="https://github.com/nvanva/nb-mlm" additional="false">nvanva/nb-mlm</pwccode>
       <video href="2021.emnlp-main.717.mp4"/>
+      <pwccode url="https://github.com/nvanva/nb-mlm" additional="false">nvanva/nb-mlm</pwccode>
     </paper>
     <paper id="718">
       <title>Revisiting Self-training for Few-shot Learning of Language Model</title>
@@ -11209,6 +11216,7 @@
       <url hash="5f86eba3">2021.emnlp-main.718</url>
       <bibkey>chen-etal-2021-revisiting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.718</doi>
+      <video href="2021.emnlp-main.718.mp4"/>
       <pwccode url="https://github.com/matthewcym/sflm" additional="false">matthewcym/sflm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
@@ -11216,7 +11224,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.718.mp4"/>
     </paper>
     <paper id="719">
       <title>Bridging Perception, Memory, and Inference through Semantic Relations</title>
@@ -11242,9 +11249,9 @@
       <url hash="cdccbf14">2021.emnlp-main.720</url>
       <bibkey>guo-etal-2021-unimodal</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.720</doi>
+      <video href="2021.emnlp-main.720.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cmu-mosei">CMU-MOSEI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multimodal-opinionlevel-sentiment-intensity">Multimodal Opinionlevel Sentiment Intensity</pwcdataset>
-      <video href="2021.emnlp-main.720.mp4"/>
     </paper>
     <paper id="721">
       <title><fixed-case>YASO</fixed-case>: <fixed-case>A</fixed-case> Targeted Sentiment Analysis Evaluation Dataset for Open-Domain Reviews</title>
@@ -11259,10 +11266,10 @@
       <url hash="6a787f6e">2021.emnlp-main.721</url>
       <bibkey>orbach-etal-2021-yaso</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.721</doi>
+      <video href="2021.emnlp-main.721.mp4"/>
       <pwccode url="https://github.com/IBM/yaso-tsa" additional="true">IBM/yaso-tsa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/yaso">YASO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.721.mp4"/>
     </paper>
     <paper id="722">
       <title>An Empirical Study on Leveraging Position Embeddings for Target-oriented Opinion Words Extraction</title>
@@ -11274,8 +11281,8 @@
       <url hash="5bbc563c">2021.emnlp-main.722</url>
       <bibkey>mensah-etal-2021-empirical</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.722</doi>
-      <pwccode url="https://github.com/samensah/encoders_towe_emnlp2021" additional="false">samensah/encoders_towe_emnlp2021</pwccode>
       <video href="2021.emnlp-main.722.mp4"/>
+      <pwccode url="https://github.com/samensah/encoders_towe_emnlp2021" additional="false">samensah/encoders_towe_emnlp2021</pwccode>
     </paper>
     <paper id="723">
       <title>Improving Multimodal Fusion with Hierarchical Mutual Information Maximization for Multimodal Sentiment Analysis</title>
@@ -11287,8 +11294,8 @@
       <url hash="b4b944a6">2021.emnlp-main.723</url>
       <bibkey>han-etal-2021-improving</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.723</doi>
-      <pwccode url="https://github.com/declare-lab/multimodal-deep-learning" additional="true">declare-lab/multimodal-deep-learning</pwccode>
       <video href="2021.emnlp-main.723.mp4"/>
+      <pwccode url="https://github.com/declare-lab/multimodal-deep-learning" additional="true">declare-lab/multimodal-deep-learning</pwccode>
     </paper>
     <paper id="724">
       <title><fixed-case>BERT</fixed-case>4<fixed-case>GCN</fixed-case>: Using <fixed-case>BERT</fixed-case> Intermediate Layers to Augment <fixed-case>GCN</fixed-case> for Aspect-based Sentiment Classification</title>
@@ -11327,8 +11334,8 @@
       <url hash="e8e78567">2021.emnlp-main.726</url>
       <bibkey>zhang-etal-2021-aspect-sentiment</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.726</doi>
-      <pwccode url="https://github.com/isakzhang/absa-quad" additional="false">isakzhang/absa-quad</pwccode>
       <video href="2021.emnlp-main.726.mp4"/>
+      <pwccode url="https://github.com/isakzhang/absa-quad" additional="false">isakzhang/absa-quad</pwccode>
     </paper>
     <paper id="727">
       <title>Cross-lingual Aspect-based Sentiment Analysis with Aspect Term Code-Switching</title>
@@ -11342,8 +11349,8 @@
       <url hash="0b5024cc">2021.emnlp-main.727</url>
       <bibkey>zhang-etal-2021-cross</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.727</doi>
-      <pwccode url="https://github.com/isakzhang/xabsa" additional="false">isakzhang/xabsa</pwccode>
       <video href="2021.emnlp-main.727.mp4"/>
+      <pwccode url="https://github.com/isakzhang/xabsa" additional="false">isakzhang/xabsa</pwccode>
     </paper>
     <paper id="728">
       <title>Towards Label-Agnostic Emotion Embeddings</title>
@@ -11368,8 +11375,8 @@
       <url hash="1dd92b9b">2021.emnlp-main.729</url>
       <bibkey>ma-etal-2021-collaborative</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.729</doi>
-      <pwccode url="https://github.com/sunlight-ym/cbd_style_transfer" additional="false">sunlight-ym/cbd_style_transfer</pwccode>
       <video href="2021.emnlp-main.729.mp4"/>
+      <pwccode url="https://github.com/sunlight-ym/cbd_style_transfer" additional="false">sunlight-ym/cbd_style_transfer</pwccode>
     </paper>
     <paper id="730">
       <title>Exploring Non-Autoregressive Text Style Transfer</title>
@@ -11380,9 +11387,9 @@
       <url hash="78b4edf6">2021.emnlp-main.730</url>
       <bibkey>ma-li-2021-exploring</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.730</doi>
+      <video href="2021.emnlp-main.730.mp4"/>
       <pwccode url="https://github.com/sunlight-ym/nar_style_transfer" additional="false">sunlight-ym/nar_style_transfer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
-      <video href="2021.emnlp-main.730.mp4"/>
     </paper>
     <paper id="731">
       <title><fixed-case>PASTE</fixed-case>: A Tagging-Free Decoding Framework Using Pointer Networks for Aspect Sentiment Triplet Extraction</title>
@@ -11396,9 +11403,9 @@
       <url hash="c219823b">2021.emnlp-main.731</url>
       <bibkey>mukherjee-etal-2021-paste</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.731</doi>
+      <video href="2021.emnlp-main.731.mp4"/>
       <pwccode url="https://github.com/rajdeep345/paste" additional="false">rajdeep345/paste</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/aste-data-v2">ASTE-Data-V2</pwcdataset>
-      <video href="2021.emnlp-main.731.mp4"/>
     </paper>
     <paper id="732">
       <title>Adaptive Proposal Generation Network for Temporal Sentence Localization in Videos</title>
@@ -11411,9 +11418,9 @@
       <url hash="87734e67">2021.emnlp-main.732</url>
       <bibkey>liu-etal-2021-adaptive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.732</doi>
+      <video href="2021.emnlp-main.732.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/activitynet-captions">ActivityNet Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/charades-sta">Charades-STA</pwcdataset>
-      <video href="2021.emnlp-main.732.mp4"/>
     </paper>
     <paper id="733">
       <title>Progressively Guide to Attend: An Iterative Alignment Framework for Temporal Sentence Grounding</title>
@@ -11425,10 +11432,10 @@
       <url hash="d7fa8de9">2021.emnlp-main.733</url>
       <bibkey>liu-etal-2021-progressively</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.733</doi>
+      <video href="2021.emnlp-main.733.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/activitynet-captions">ActivityNet Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/charades">Charades</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/charades-sta">Charades-STA</pwcdataset>
-      <video href="2021.emnlp-main.733.mp4"/>
     </paper>
     <paper id="734">
       <title>Language Models are Few-Shot Butlers</title>
@@ -11439,8 +11446,8 @@
       <url hash="433dab29">2021.emnlp-main.734</url>
       <bibkey>micheli-fleuret-2021-language</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.734</doi>
-      <pwccode url="https://github.com/vmicheli/lm-butlers" additional="false">vmicheli/lm-butlers</pwccode>
       <video href="2021.emnlp-main.734.mp4"/>
+      <pwccode url="https://github.com/vmicheli/lm-butlers" additional="false">vmicheli/lm-butlers</pwccode>
     </paper>
     <paper id="735">
       <title><fixed-case>R</fixed-case>ˆ3<fixed-case>N</fixed-case>et:Relation-embedded Representation Reconstruction Network for Change Captioning</title>
@@ -11454,9 +11461,9 @@
       <url hash="57e03dc5">2021.emnlp-main.735</url>
       <bibkey>tu-etal-2021-r</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.735</doi>
+      <video href="2021.emnlp-main.735.mp4"/>
       <pwccode url="https://github.com/tuyunbin/r3net" additional="false">tuyunbin/r3net</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/spot-the-diff">Spot-the-diff</pwcdataset>
-      <video href="2021.emnlp-main.735.mp4"/>
     </paper>
     <paper id="736">
       <title>Looking for Confirmations: An Effective and Human-Like Visual Dialogue Strategy</title>
@@ -11467,9 +11474,9 @@
       <url hash="c24b5e9c">2021.emnlp-main.736</url>
       <bibkey>testoni-bernardi-2021-looking</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.736</doi>
+      <video href="2021.emnlp-main.736.mp4"/>
       <pwccode url="https://github.com/albertotestoni/confirm_it" additional="false">albertotestoni/confirm_it</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/guesswhat">GuessWhat?!</pwcdataset>
-      <video href="2021.emnlp-main.736.mp4"/>
     </paper>
     <paper id="737">
       <title>A Unified Speaker Adaptation Approach for <fixed-case>ASR</fixed-case></title>
@@ -11484,9 +11491,9 @@
       <url hash="5270b09d">2021.emnlp-main.737</url>
       <bibkey>zhao-etal-2021-unified</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.737</doi>
+      <video href="2021.emnlp-main.737.mp4"/>
       <pwccode url="https://github.com/zyzpower/gradprune_speaker" additional="false">zyzpower/gradprune_speaker</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
-      <video href="2021.emnlp-main.737.mp4"/>
     </paper>
     <paper id="738">
       <title>Caption Enriched Samples for Improving Hateful Memes Detection</title>
@@ -11499,8 +11506,8 @@
       <attachment type="Software" hash="1a72fda5">2021.emnlp-main.738.Software.zip</attachment>
       <bibkey>blaier-etal-2021-caption</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.738</doi>
-      <pwccode url="https://github.com/efrat-safanov/caption-enriched-samples-research" additional="false">efrat-safanov/caption-enriched-samples-research</pwccode>
       <video href="2021.emnlp-main.738.mp4"/>
+      <pwccode url="https://github.com/efrat-safanov/caption-enriched-samples-research" additional="false">efrat-safanov/caption-enriched-samples-research</pwccode>
     </paper>
     <paper id="739">
       <title>Sparsity and Sentence Structure in Encoder-Decoder Attention of Summarization Systems</title>
@@ -11523,11 +11530,11 @@
       <url hash="afdbcc87">2021.emnlp-main.740</url>
       <bibkey>kamal-eddine-etal-2021-barthez</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.740</doi>
+      <video href="2021.emnlp-main.740.mp4"/>
       <pwccode url="https://github.com/moussaKam/BARThez" additional="true">moussaKam/BARThez</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/orangesum">OrangeSum</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flue-french-language-understanding-evaluation">FLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
-      <video href="2021.emnlp-main.740.mp4"/>
     </paper>
     <paper id="741">
       <title><fixed-case>ARMAN</fixed-case>: <fixed-case>P</fixed-case>re-training with <fixed-case>S</fixed-case>emantically <fixed-case>S</fixed-case>electing and <fixed-case>R</fixed-case>eordering of <fixed-case>S</fixed-case>entences for <fixed-case>P</fixed-case>ersian <fixed-case>A</fixed-case>bstractive <fixed-case>S</fixed-case>ummarization</title>
@@ -11540,11 +11547,11 @@
       <url hash="9e52027f">2021.emnlp-main.741</url>
       <bibkey>salemi-etal-2021-arman</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.741</doi>
+      <video href="2021.emnlp-main.741.mp4"/>
       <pwccode url="https://github.com/alirezasalemi7/arman" additional="false">alirezasalemi7/arman</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cc100">CC100</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/perkey">PerKey</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/pn-summary">pn-summary</pwcdataset>
-      <video href="2021.emnlp-main.741.mp4"/>
     </paper>
     <paper id="742">
       <title>Models and Datasets for Cross-Lingual Summarisation</title>
@@ -11555,9 +11562,9 @@
       <url hash="9222eb1e">2021.emnlp-main.742</url>
       <bibkey>perez-beltrachini-lapata-2021-models</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.742</doi>
+      <video href="2021.emnlp-main.742.mp4"/>
       <pwccode url="https://github.com/lauhaide/clads" additional="false">lauhaide/clads</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikilingua">WikiLingua</pwcdataset>
-      <video href="2021.emnlp-main.742.mp4"/>
     </paper>
     <paper id="743">
       <title>Learning Opinion Summarizers by Selecting Informative Reviews</title>
@@ -11569,8 +11576,8 @@
       <url hash="e5c8a7b5">2021.emnlp-main.743</url>
       <bibkey>brazinskas-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.743</doi>
-      <pwccode url="https://github.com/abrazinskas/selsum" additional="false">abrazinskas/selsum</pwccode>
       <video href="2021.emnlp-main.743.mp4"/>
+      <pwccode url="https://github.com/abrazinskas/selsum" additional="false">abrazinskas/selsum</pwccode>
     </paper>
     <paper id="744">
       <title>Enriching and Controlling Global Semantics for Text Summarization</title>
@@ -11583,8 +11590,8 @@
       <url hash="62f529cc">2021.emnlp-main.744</url>
       <bibkey>nguyen-etal-2021-enriching</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.744</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/reddit-tifu">Reddit TIFU</pwcdataset>
       <video href="2021.emnlp-main.744.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/reddit-tifu">Reddit TIFU</pwcdataset>
     </paper>
     <paper id="745">
       <title>Revisiting Tri-training of Dependency Parsers</title>
@@ -11595,8 +11602,8 @@
       <url hash="e899d7c6">2021.emnlp-main.745</url>
       <bibkey>wagner-foster-2021-revisiting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.745</doi>
-      <pwccode url="https://github.com/jowagner/mtb-tri-training" additional="true">jowagner/mtb-tri-training</pwccode>
       <video href="2021.emnlp-main.745.mp4"/>
+      <pwccode url="https://github.com/jowagner/mtb-tri-training" additional="true">jowagner/mtb-tri-training</pwccode>
     </paper>
     <paper id="746">
       <title>Bridge to Target Domain by Prototypical Contrastive Learning and Label Confusion: Re-explore Zero-Shot Learning for Slot Filling</title>
@@ -11611,8 +11618,8 @@
       <url hash="5075be54">2021.emnlp-main.746</url>
       <bibkey>wang-etal-2021-bridge</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.746</doi>
-      <pwccode url="https://github.com/w-lw/pclc" additional="false">w-lw/pclc</pwccode>
       <video href="2021.emnlp-main.746.mp4"/>
+      <pwccode url="https://github.com/w-lw/pclc" additional="false">w-lw/pclc</pwccode>
     </paper>
     <paper id="747">
       <title>Neuralizing Regular Expressions for Slot Filling</title>
@@ -11624,8 +11631,8 @@
       <url hash="07485a04">2021.emnlp-main.747</url>
       <bibkey>jiang-etal-2021-neuralizing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.747</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <video href="2021.emnlp-main.747.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
     </paper>
     <paper id="748">
       <title>Causal Direction of Data Collection Matters: Implications of Causal and Anticausal Learning for <fixed-case>NLP</fixed-case></title>
@@ -11641,8 +11648,8 @@
       <url hash="71557b46">2021.emnlp-main.748</url>
       <bibkey>jin-etal-2021-causal</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.748</doi>
-      <pwccode url="https://github.com/zhijing-jin/icm4nlp" additional="false">zhijing-jin/icm4nlp</pwccode>
       <video href="2021.emnlp-main.748.mp4"/>
+      <pwccode url="https://github.com/zhijing-jin/icm4nlp" additional="false">zhijing-jin/icm4nlp</pwccode>
     </paper>
     <paper id="749">
       <title>Raise a Child in Large Language Model: Towards Effective and Generalizable Fine-tuning</title>
@@ -11658,12 +11665,12 @@
       <url hash="0417c456">2021.emnlp-main.749</url>
       <bibkey>xu-etal-2021-raise</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.749</doi>
+      <video href="2021.emnlp-main.749.mp4"/>
       <pwccode url="https://github.com/alibaba/AliceMind" additional="true">alibaba/AliceMind</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sick">SICK</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.emnlp-main.749.mp4"/>
     </paper>
     <paper id="750">
       <title>Knowledge Graph Representation Learning using Ordinary Differential Equations</title>
@@ -11678,8 +11685,8 @@
       <url hash="c81fa505">2021.emnlp-main.750</url>
       <bibkey>nayyeri-etal-2021-knowledge</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.750</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
       <video href="2021.emnlp-main.750.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
     </paper>
     <paper id="751">
       <title><fixed-case>K</fixed-case>now<fixed-case>MAN</fixed-case>: Weakly Supervised Multinomial Adversarial Networks</title>
@@ -11693,8 +11700,8 @@
       <url hash="c439f29a">2021.emnlp-main.751</url>
       <bibkey>marz-etal-2021-knowman</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.751</doi>
-      <pwccode url="https://github.com/luisamaerz/knowman" additional="false">luisamaerz/knowman</pwccode>
       <video href="2021.emnlp-main.751.mp4"/>
+      <pwccode url="https://github.com/luisamaerz/knowman" additional="false">luisamaerz/knowman</pwccode>
     </paper>
     <paper id="752">
       <title><fixed-case>ONION</fixed-case>: A Simple and Effective Defense Against Textual Backdoor Attacks</title>
@@ -11709,10 +11716,10 @@
       <url hash="1b8e14d1">2021.emnlp-main.752</url>
       <bibkey>qi-etal-2021-onion</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.752</doi>
+      <video href="2021.emnlp-main.752.mp4"/>
       <pwccode url="https://github.com/thunlp/ONION" additional="true">thunlp/ONION</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.752.mp4"/>
     </paper>
     <paper id="753">
       <title>Value-aware Approximate Attention</title>
@@ -11723,8 +11730,8 @@
       <url hash="5681b8c8">2021.emnlp-main.753</url>
       <bibkey>gupta-berant-2021-value</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.753</doi>
-      <pwccode url="https://github.com/ag1988/value_aware_attn" additional="false">ag1988/value_aware_attn</pwccode>
       <video href="2021.emnlp-main.753.mp4"/>
+      <pwccode url="https://github.com/ag1988/value_aware_attn" additional="false">ag1988/value_aware_attn</pwccode>
     </paper>
     <paper id="754">
       <title>Contrastive Domain Adaptation for Question Answering using Limited Text Corpora</title>
@@ -11736,13 +11743,13 @@
       <url hash="65cbf6e1">2021.emnlp-main.754</url>
       <bibkey>yue-etal-2021-contrastive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.754</doi>
+      <video href="2021.emnlp-main.754.mp4"/>
       <pwccode url="https://github.com/yueeeeeeee/caqa" additional="false">yueeeeeeee/caqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/searchqa">SearchQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.emnlp-main.754.mp4"/>
     </paper>
     <paper id="755">
       <title>Case-based Reasoning for Natural Language Queries over Knowledge Bases</title>
@@ -11760,12 +11767,12 @@
       <url hash="c8a8afde">2021.emnlp-main.755</url>
       <bibkey>das-etal-2021-case</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.755</doi>
+      <video href="2021.emnlp-main.755.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/complexwebquestions">ComplexWebQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/simplequestions">SimpleQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestionssp">WebQuestionsSP</pwcdataset>
-      <video href="2021.emnlp-main.755.mp4"/>
     </paper>
     <paper id="756">
       <title>Distantly-Supervised Dense Retrieval Enables Open-Domain Question Answering without Evidence Annotation</title>
@@ -11778,8 +11785,8 @@
       <url hash="f69c33b4">2021.emnlp-main.756</url>
       <bibkey>zhao-etal-2021-distantly</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.756</doi>
-      <pwccode url="https://github.com/henryzhao5852/distdr" additional="false">henryzhao5852/distdr</pwccode>
       <video href="2021.emnlp-main.756.mp4"/>
+      <pwccode url="https://github.com/henryzhao5852/distdr" additional="false">henryzhao5852/distdr</pwccode>
     </paper>
     <paper id="757">
       <title>What’s in a Name? Answer Equivalence For Open-Domain Question Answering</title>
@@ -11805,8 +11812,8 @@
       <url hash="70b52233">2021.emnlp-main.758</url>
       <bibkey>rodriguez-boyd-graber-2021-evaluation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.758</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <video href="2021.emnlp-main.758.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
     </paper>
     <paper id="759">
       <title>Numerical reasoning in machine reading comprehension tasks: are we there yet?</title>
@@ -11818,8 +11825,8 @@
       <url hash="71bbd697">2021.emnlp-main.759</url>
       <bibkey>al-negheimish-etal-2021-numerical</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.759</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <video href="2021.emnlp-main.759.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
     </paper>
     <paper id="760">
       <title>Set Generation Networks for End-to-End Knowledge Base Population</title>
@@ -11846,8 +11853,8 @@
       <url hash="fd9f0c38">2021.emnlp-main.761</url>
       <bibkey>hao-etal-2021-knowing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.761</doi>
-      <pwccode url="https://github.com/nju-websoft/fan" additional="false">nju-websoft/fan</pwccode>
       <video href="2021.emnlp-main.761.mp4"/>
+      <pwccode url="https://github.com/nju-websoft/fan" additional="false">nju-websoft/fan</pwccode>
     </paper>
     <paper id="762">
       <title>Progressive Adversarial Learning for Bootstrapping: A Case Study on Entity Set Expansion</title>
@@ -11859,8 +11866,8 @@
       <url hash="a739d6d0">2021.emnlp-main.762</url>
       <bibkey>yan-etal-2021-progressive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.762</doi>
-      <pwccode url="https://github.com/lingyongyan/bootstrapgan" additional="false">lingyongyan/bootstrapgan</pwccode>
       <video href="2021.emnlp-main.762.mp4"/>
+      <pwccode url="https://github.com/lingyongyan/bootstrapgan" additional="false">lingyongyan/bootstrapgan</pwccode>
     </paper>
     <paper id="763">
       <title>Uncovering Main Causalities for Long-tailed Information Extraction</title>
@@ -11874,9 +11881,9 @@
       <url hash="05dfecf0">2021.emnlp-main.763</url>
       <bibkey>nan-etal-2021-uncovering</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.763</doi>
+      <video href="2021.emnlp-main.763.mp4"/>
       <pwccode url="https://github.com/heyyyyyyg/cfie" additional="false">heyyyyyyg/cfie</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/maven">MAVEN</pwcdataset>
-      <video href="2021.emnlp-main.763.mp4"/>
     </paper>
     <paper id="764">
       <title>Maximal Clique Based Non-Autoregressive Open Information Extraction</title>
@@ -11904,8 +11911,8 @@
       <url hash="fb4e65b2">2021.emnlp-main.765</url>
       <bibkey>zhao-etal-2021-relation</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.765</doi>
-      <pwccode url="https://github.com/ac-zyx/rocore" additional="false">ac-zyx/rocore</pwccode>
       <video href="2021.emnlp-main.765.mp4"/>
+      <pwccode url="https://github.com/ac-zyx/rocore" additional="false">ac-zyx/rocore</pwccode>
     </paper>
     <paper id="766">
       <title>Exploring Methods for Generating Feedback Comments for Writing Learning</title>
@@ -11917,8 +11924,8 @@
       <url hash="28a61cb0">2021.emnlp-main.766</url>
       <bibkey>hanawa-etal-2021-exploring</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.766</doi>
-      <pwccode url="https://github.com/k-hanawa/fcg_emnlp2021" additional="false">k-hanawa/fcg_emnlp2021</pwccode>
       <video href="2021.emnlp-main.766.mp4"/>
+      <pwccode url="https://github.com/k-hanawa/fcg_emnlp2021" additional="false">k-hanawa/fcg_emnlp2021</pwccode>
     </paper>
     <paper id="767">
       <title>A Role-Selected Sharing Network for Joint Machine-Human Chatting Handoff and Service Satisfaction Analysis</title>
@@ -11935,8 +11942,8 @@
       <url hash="b9213d0d">2021.emnlp-main.767</url>
       <bibkey>liu-etal-2021-role</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.767</doi>
-      <pwccode url="https://github.com/weijialau/rssn" additional="false">weijialau/rssn</pwccode>
       <video href="2021.emnlp-main.767.mp4"/>
+      <pwccode url="https://github.com/weijialau/rssn" additional="false">weijialau/rssn</pwccode>
     </paper>
     <paper id="768">
       <title>Meta Distant Transfer Learning for Pre-trained Language Models</title>
@@ -11951,9 +11958,9 @@
       <url hash="86cfed8c">2021.emnlp-main.768</url>
       <bibkey>wang-etal-2021-meta-distant</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.768</doi>
+      <video href="2021.emnlp-main.768.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.768.mp4"/>
     </paper>
     <paper id="769">
       <title><fixed-case>U</fixed-case>ni<fixed-case>KER</fixed-case>: A Unified Framework for Combining Embedding and Definite Horn Rule Reasoning for Knowledge Graph Inference</title>
@@ -12010,8 +12017,8 @@
       <attachment type="Software" hash="52e60ac5">2021.emnlp-main.772.Software.txt</attachment>
       <bibkey>liu-etal-2021-inflate</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.772</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
       <video href="2021.emnlp-main.772.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
     </paper>
     <paper id="773">
       <title>On Pursuit of Designing Multi-modal Transformer for Video Grounding</title>
@@ -12027,9 +12034,9 @@
       <doi>10.18653/v1/2021.emnlp-main.773</doi>
       <revision id="1" href="2021.emnlp-main.773v1" hash="2bd6c814"/>
       <revision id="2" href="2021.emnlp-main.773v2" hash="67076422" date="2022-04-29">Added missing acknowledgement.</revision>
+      <video href="2021.emnlp-main.773.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/activitynet-captions">ActivityNet Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/charades-sta">Charades-STA</pwcdataset>
-      <video href="2021.emnlp-main.773.mp4"/>
     </paper>
     <paper id="774">
       <title><fixed-case>COVR</fixed-case>: A Test-Bed for Visually Grounded Compositional Generalization with Real Images</title>
@@ -12042,10 +12049,10 @@
       <url hash="1fe9e067">2021.emnlp-main.774</url>
       <bibkey>bogin-etal-2021-covr</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.774</doi>
+      <video href="2021.emnlp-main.774.mp4"/>
       <pwccode url="https://github.com/benbogin/covr-dataset" additional="false">benbogin/covr-dataset</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.emnlp-main.774.mp4"/>
     </paper>
     <paper id="775">
       <title>Vision-and-Language or Vision-for-Language? On Cross-Modal Influence in Multimodal Transformers</title>
@@ -12057,10 +12064,10 @@
       <url hash="4105ff84">2021.emnlp-main.775</url>
       <bibkey>frank-etal-2021-vision</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.775</doi>
+      <video href="2021.emnlp-main.775.mp4"/>
       <pwccode url="https://github.com/e-bug/volta" additional="true">e-bug/volta</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
-      <video href="2021.emnlp-main.775.mp4"/>
     </paper>
     <paper id="776">
       <title><fixed-case>H</fixed-case>yp<fixed-case>M</fixed-case>ix: Hyperbolic Interpolative Data Augmentation</title>
@@ -12075,10 +12082,10 @@
       <url hash="893a4936">2021.emnlp-main.776</url>
       <bibkey>sawhney-etal-2021-hypmix</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.776</doi>
+      <video href="2021.emnlp-main.776.mp4"/>
       <pwccode url="https://github.com/caisa-lab/hypmix-emnlp" additional="false">caisa-lab/hypmix-emnlp</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cifar-10">CIFAR-10</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cifar-100">CIFAR-100</pwcdataset>
-      <video href="2021.emnlp-main.776.mp4"/>
     </paper>
     <paper id="777">
       <title>Integrating Deep Event-Level and Script-Level Information for Script Event Prediction</title>
@@ -12093,8 +12100,8 @@
       <url hash="1df22a15">2021.emnlp-main.777</url>
       <bibkey>bai-etal-2021-integrating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.777</doi>
-      <pwccode url="https://github.com/waltbai/MCPredictor" additional="false">waltbai/MCPredictor</pwccode>
       <video href="2021.emnlp-main.777.mp4"/>
+      <pwccode url="https://github.com/waltbai/MCPredictor" additional="false">waltbai/MCPredictor</pwccode>
     </paper>
     <paper id="778">
       <title><fixed-case>QA</fixed-case>-Align: Representing Cross-Text Content Overlap by Aligning Question-Answer Propositions</title>
@@ -12108,10 +12115,10 @@
       <url hash="6bd2ac49">2021.emnlp-main.778</url>
       <bibkey>brook-weiss-etal-2021-qa</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.778</doi>
+      <video href="2021.emnlp-main.778.mp4"/>
       <pwccode url="https://github.com/danielabweiss/qa-align" additional="false">danielabweiss/qa-align</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ecb">ECB+</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qa-srl">QA-SRL</pwcdataset>
-      <video href="2021.emnlp-main.778.mp4"/>
     </paper>
     <paper id="779">
       <title><fixed-case>PICARD</fixed-case>: Parsing Incrementally for Constrained Auto-Regressive Decoding from Language Models</title>
@@ -12123,10 +12130,10 @@
       <url hash="6b45cca4">2021.emnlp-main.779</url>
       <bibkey>scholak-etal-2021-picard</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.779</doi>
+      <video href="2021.emnlp-main.779.mp4"/>
       <pwccode url="https://github.com/ElementAI/picard" additional="false">ElementAI/picard</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cosql">CoSQL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
-      <video href="2021.emnlp-main.779.mp4"/>
     </paper>
     <paper id="780">
       <title>Exploiting <fixed-case>T</fixed-case>witter as Source of Large Corpora of Weakly Similar Pairs for Semantic Sentence Embeddings</title>
@@ -12137,10 +12144,10 @@
       <url hash="fcbf27ea">2021.emnlp-main.780</url>
       <bibkey>di-giovanni-brambilla-2021-exploiting</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.780</doi>
+      <video href="2021.emnlp-main.780.mp4"/>
       <pwccode url="https://github.com/marco-digio/twitter4sse" additional="false">marco-digio/twitter4sse</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/pit">PIT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/twitter-news-url-corpus">TURL</pwcdataset>
-      <video href="2021.emnlp-main.780.mp4"/>
     </paper>
     <paper id="781">
       <title>Guilt by Association: Emotion Intensities in Lexical Representations</title>
@@ -12173,8 +12180,8 @@
       <url hash="142d81b0">2021.emnlp-main.783</url>
       <bibkey>roy-etal-2021-identifying</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.783</doi>
-      <pwccode url="https://github.com/shamikroy/moral-role-prediction" additional="false">shamikroy/moral-role-prediction</pwccode>
       <video href="2021.emnlp-main.783.mp4"/>
+      <pwccode url="https://github.com/shamikroy/moral-role-prediction" additional="false">shamikroy/moral-role-prediction</pwccode>
     </paper>
     <paper id="784">
       <title>Measuring Sentence-Level and Aspect-Level (Un)certainty in Science Communications</title>
@@ -12197,8 +12204,9 @@
       <url hash="1c32913d">2021.emnlp-main.785</url>
       <bibkey>du-etal-2021-assessing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.785</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
       <video href="2021.emnlp-main.785.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="786">
       <title>Rumor Detection on <fixed-case>T</fixed-case>witter with Claim-Guided Hierarchical Graph Attention Networks</title>
@@ -12227,8 +12235,8 @@
       <attachment type="Software" hash="bfe0a9b4">2021.emnlp-main.787.Software.zip</attachment>
       <bibkey>kim-etal-2021-learning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.787</doi>
-      <pwccode url="https://github.com/hikoseon12/learning-bill-similarity" additional="false">hikoseon12/learning-bill-similarity</pwccode>
       <video href="2021.emnlp-main.787.mp4"/>
+      <pwccode url="https://github.com/hikoseon12/learning-bill-similarity" additional="false">hikoseon12/learning-bill-similarity</pwccode>
     </paper>
     <paper id="788">
       <title><fixed-case>SWEAT</fixed-case>: Scoring Polarization of Topics across Different Corpora</title>
@@ -12241,8 +12249,8 @@
       <url hash="7b9953e7">2021.emnlp-main.788</url>
       <bibkey>bianchi-etal-2021-sweat</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.788</doi>
-      <pwccode url="https://github.com/vinid/sweat" additional="false">vinid/sweat</pwccode>
       <video href="2021.emnlp-main.788.mp4"/>
+      <pwccode url="https://github.com/vinid/sweat" additional="false">vinid/sweat</pwccode>
     </paper>
     <paper id="789">
       <title>“So You Think You’re Funny?”: Rating the Humour Quotient in Standup Comedy</title>
@@ -12256,8 +12264,8 @@
       <url hash="06308fa1">2021.emnlp-main.789</url>
       <bibkey>mittal-etal-2021-think</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.789</doi>
-      <pwccode url="https://github.com/theextrasemicolon/ai-openmic" additional="false">theextrasemicolon/ai-openmic</pwccode>
       <video href="2021.emnlp-main.789.mp4"/>
+      <pwccode url="https://github.com/theextrasemicolon/ai-openmic" additional="false">theextrasemicolon/ai-openmic</pwccode>
     </paper>
     <paper id="790">
       <title>“Was it “stated” or was it “claimed”?: How linguistic bias affects generative language models</title>
@@ -12282,12 +12290,12 @@
       <url hash="51096494">2021.emnlp-main.791</url>
       <bibkey>cao-etal-2021-pause</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.791</doi>
+      <video href="2021.emnlp-main.791.mp4"/>
       <pwccode url="https://github.com/eqtpartners/pause" additional="false">eqtpartners/pause</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.emnlp-main.791.mp4"/>
     </paper>
     <paper id="792">
       <title>A Simple Geometric Method for Cross-Lingual Linguistic Transformations with Pre-trained Autoencoders</title>
@@ -12301,8 +12309,8 @@
       <url hash="c9568ff6">2021.emnlp-main.792</url>
       <bibkey>de-raedt-etal-2021-simple</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.792</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
       <video href="2021.emnlp-main.792.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
     </paper>
     <paper id="793">
       <title>An Information-Theoretic Characterization of Morphological Fusion</title>
@@ -12314,8 +12322,8 @@
       <url hash="3ffc2dba">2021.emnlp-main.793</url>
       <bibkey>rathi-etal-2021-information</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.793</doi>
-      <pwccode url="https://github.com/neilrathi/morphological-fusion" additional="false">neilrathi/morphological-fusion</pwccode>
       <video href="2021.emnlp-main.793.mp4"/>
+      <pwccode url="https://github.com/neilrathi/morphological-fusion" additional="false">neilrathi/morphological-fusion</pwccode>
     </paper>
     <paper id="794">
       <title>The Effect of Efficient Messaging and Input Variability on Neural-Agent Iterated Language Learning</title>
@@ -12341,8 +12349,8 @@
       <url hash="d8c53cf2">2021.emnlp-main.795</url>
       <bibkey>korner-etal-2021-classifying</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.795</doi>
-      <pwccode url="https://github.com/webis-de/emnlp-21" additional="false">webis-de/emnlp-21</pwccode>
       <video href="2021.emnlp-main.795.mp4"/>
+      <pwccode url="https://github.com/webis-de/emnlp-21" additional="false">webis-de/emnlp-21</pwccode>
     </paper>
     <paper id="796">
       <title><fixed-case>C</fixed-case>hinese Opinion Role Labeling with Corpus Translation: A Pivot Study</title>
@@ -12357,9 +12365,9 @@
       <attachment type="Software" hash="f67c6879">2021.emnlp-main.796.Software.zip</attachment>
       <bibkey>zhen-etal-2021-chinese</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.796</doi>
+      <video href="2021.emnlp-main.796.mp4"/>
       <pwccode url="https://github.com/zenrran/chineseorl-with-corpus-translation" additional="false">zenrran/chineseorl-with-corpus-translation</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
-      <video href="2021.emnlp-main.796.mp4"/>
     </paper>
     <paper id="797">
       <title><fixed-case>M</fixed-case>assive<fixed-case>S</fixed-case>umm: a very large-scale, very multilingual, news summarisation dataset</title>
@@ -12370,13 +12378,13 @@
       <url hash="f78971c7">2021.emnlp-main.797</url>
       <bibkey>varab-schluter-2021-massivesumm</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.797</doi>
+      <video href="2021.emnlp-main.797.mp4"/>
       <pwccode url="https://github.com/danielvarab/massive-summ" additional="false">danielvarab/massive-summ</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/global-voices">Global Voices</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mlsum">MLSUM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsroom">NEWSROOM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/new-york-times-annotated-corpus">New York Times Annotated Corpus</pwcdataset>
-      <video href="2021.emnlp-main.797.mp4"/>
     </paper>
     <paper id="798">
       <title><fixed-case>AUTOSUMM</fixed-case>: Automatic Model Creation for Text Summarization</title>
@@ -12406,8 +12414,8 @@
       <attachment type="Software" hash="9178ada6">2021.emnlp-main.799.Software.zip</attachment>
       <bibkey>shenoy-etal-2021-investigating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.799</doi>
-      <pwccode url="https://github.com/nicoherbig/mmpe" additional="false">nicoherbig/mmpe</pwccode>
       <video href="2021.emnlp-main.799.mp4"/>
+      <pwccode url="https://github.com/nicoherbig/mmpe" additional="false">nicoherbig/mmpe</pwccode>
     </paper>
     <paper id="800">
       <title><fixed-case>UNK</fixed-case>s Everywhere: <fixed-case>A</fixed-case>dapting Multilingual Language Models to New Scripts</title>
@@ -12420,9 +12428,9 @@
       <url hash="9adf6cc2">2021.emnlp-main.800</url>
       <bibkey>pfeiffer-etal-2021-unks</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.800</doi>
+      <video href="2021.emnlp-main.800.mp4"/>
       <pwccode url="https://github.com/adapter-hub/unks_everywhere" additional="true">adapter-hub/unks_everywhere</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.emnlp-main.800.mp4"/>
     </paper>
     <paper id="801">
       <title>Neural Machine Translation Quality and Post-Editing Performance</title>
@@ -12456,6 +12464,7 @@
       <url hash="1ba70470">2021.emnlp-main.802</url>
       <bibkey>ruder-etal-2021-xtreme</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.802</doi>
+      <video href="2021.emnlp-main.802.mp4"/>
       <pwccode url="https://github.com/google-research/xtreme" additional="false">google-research/xtreme</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad-r">LAReQA</pwcdataset>
@@ -12469,7 +12478,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/xcopa">XCOPA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.emnlp-main.802.mp4"/>
     </paper>
     <paper id="803">
       <title>Contrastive Conditioning for Assessing Disambiguation in <fixed-case>MT</fixed-case>: <fixed-case>A</fixed-case> Case Study of Distilled Bias</title>
@@ -12480,8 +12488,8 @@
       <url hash="386824f8">2021.emnlp-main.803</url>
       <bibkey>vamvas-sennrich-2021-contrastive</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.803</doi>
-      <pwccode url="https://github.com/zurichnlp/contrastive-conditioning" additional="false">zurichnlp/contrastive-conditioning</pwccode>
       <video href="2021.emnlp-main.803.mp4"/>
+      <pwccode url="https://github.com/zurichnlp/contrastive-conditioning" additional="false">zurichnlp/contrastive-conditioning</pwccode>
     </paper>
     <paper id="804">
       <title><fixed-case>M</fixed-case>easuring Association Between Labels and Free-Text Rationales</title>
@@ -12493,12 +12501,12 @@
       <url hash="4876a357">2021.emnlp-main.804</url>
       <bibkey>wiegreffe-etal-2021-measuring</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.804</doi>
+      <video href="2021.emnlp-main.804.mp4"/>
       <pwccode url="https://github.com/allenai/label_rationale_association" additional="false">allenai/label_rationale_association</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cos-e">CoS-E</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/commonsenseqa">CommonsenseQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/e-snli">e-SNLI</pwcdataset>
-      <video href="2021.emnlp-main.804.mp4"/>
     </paper>
     <paper id="805">
       <title>Discretized Integrated Gradients for Explaining Language Models</title>
@@ -12509,10 +12517,10 @@
       <url hash="7ad89942">2021.emnlp-main.805</url>
       <bibkey>sanyal-ren-2021-discretized</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.805</doi>
+      <video href="2021.emnlp-main.805.mp4"/>
       <pwccode url="https://github.com/ink-usc/dig" additional="false">ink-usc/dig</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.805.mp4"/>
     </paper>
     <paper id="806">
       <title>Putting Words in <fixed-case>BERT</fixed-case>’s Mouth: Navigating Contextualized Vector Spaces with Pseudowords</title>
@@ -12539,9 +12547,9 @@
       <url hash="41a62020">2021.emnlp-main.807</url>
       <bibkey>vafa-etal-2021-rationales</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.807</doi>
+      <video href="2021.emnlp-main.807.mp4"/>
       <pwccode url="https://github.com/keyonvafa/sequential-rationales" additional="true">keyonvafa/sequential-rationales</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/lambada">LAMBADA</pwcdataset>
-      <video href="2021.emnlp-main.807.mp4"/>
     </paper>
     <paper id="808">
       <title><fixed-case>F</fixed-case>ast<fixed-case>IF</fixed-case>: Scalable Influence Functions for Efficient Model Interpretation and Debugging</title>
@@ -12555,11 +12563,11 @@
       <url hash="fc43cb3a">2021.emnlp-main.808</url>
       <bibkey>guo-etal-2021-fastif</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.808</doi>
+      <video href="2021.emnlp-main.808.mp4"/>
       <pwccode url="https://github.com/salesforce/fast-influence-functions" additional="false">salesforce/fast-influence-functions</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wilds">Wilds</pwcdataset>
-      <video href="2021.emnlp-main.808.mp4"/>
     </paper>
     <paper id="809">
       <title>Studying word order through iterative shuffling</title>
@@ -12572,10 +12580,10 @@
       <url hash="c0bd06c9">2021.emnlp-main.809</url>
       <bibkey>malkin-etal-2021-studying</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.809</doi>
+      <video href="2021.emnlp-main.809.mp4"/>
       <pwccode url="https://github.com/malkin1729/ibis" additional="false">malkin1729/ibis</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.emnlp-main.809.mp4"/>
     </paper>
     <paper id="810">
       <title>Distantly-Supervised Named Entity Recognition with Noise-Robust Learning and Language Model Augmented Self-Training</title>
@@ -12591,8 +12599,8 @@
       <url hash="ed4d4e85">2021.emnlp-main.810</url>
       <bibkey>meng-etal-2021-distantly</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.810</doi>
-      <pwccode url="https://github.com/yumeng5/roster" additional="false">yumeng5/roster</pwccode>
       <video href="2021.emnlp-main.810.mp4"/>
+      <pwccode url="https://github.com/yumeng5/roster" additional="false">yumeng5/roster</pwccode>
     </paper>
     <paper id="811">
       <title>Open Knowledge Graphs Canonicalization using Variational Autoencoders</title>
@@ -12606,8 +12614,8 @@
       <url hash="339a40ac">2021.emnlp-main.811</url>
       <bibkey>dash-etal-2021-open</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.811</doi>
-      <pwccode url="https://github.com/IBM/Open-KG-canonicalization" additional="false">IBM/Open-KG-canonicalization</pwccode>
       <video href="2021.emnlp-main.811.mp4"/>
+      <pwccode url="https://github.com/IBM/Open-KG-canonicalization" additional="false">IBM/Open-KG-canonicalization</pwccode>
     </paper>
     <paper id="812">
       <title><fixed-case>H</fixed-case>itt<fixed-case>ER</fixed-case>: Hierarchical Transformers for Knowledge Graph Embeddings</title>
@@ -12622,8 +12630,8 @@
       <url hash="0ac9e4fe">2021.emnlp-main.812</url>
       <bibkey>chen-etal-2021-hitter</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.812</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
       <video href="2021.emnlp-main.812.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
     </paper>
     <paper id="813">
       <title>Few-Shot Named Entity Recognition: An Empirical Baseline Study</title>
@@ -12641,10 +12649,10 @@
       <url hash="57d06afa">2021.emnlp-main.813</url>
       <bibkey>huang-etal-2021-shot</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.813</doi>
+      <video href="2021.emnlp-main.813.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wnut-2017-emerging-and-rare-entity">WNUT 2017</pwcdataset>
-      <video href="2021.emnlp-main.813.mp4"/>
     </paper>
     <paper id="814">
       <title><fixed-case>XLE</fixed-case>nt: Mining a Large Cross-lingual Entity Dataset with Lexical-Semantic-Phonetic Word Alignment</title>
@@ -12658,8 +12666,8 @@
       <url hash="26f17252">2021.emnlp-main.814</url>
       <bibkey>el-kishky-etal-2021-xlent</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.814</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/xlent">XLEnt</pwcdataset>
       <video href="2021.emnlp-main.814.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/xlent">XLEnt</pwcdataset>
     </paper>
     <paper id="815">
       <title>Utilizing Relative Event Time to Enhance Event-Event Temporal Relation Extraction</title>
@@ -12670,8 +12678,8 @@
       <url hash="39419d19">2021.emnlp-main.815</url>
       <bibkey>wen-ji-2021-utilizing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.815</doi>
-      <pwccode url="https://github.com/wenhycs/emnlp2021-utilizing-relative-event-time-to-enhance-event-event-temporal-relation-extraction" additional="false">wenhycs/emnlp2021-utilizing-relative-event-time-to-enhance-event-event-temporal-relation-extraction</pwccode>
       <video href="2021.emnlp-main.815.mp4"/>
+      <pwccode url="https://github.com/wenhycs/emnlp2021-utilizing-relative-event-time-to-enhance-event-event-temporal-relation-extraction" additional="false">wenhycs/emnlp2021-utilizing-relative-event-time-to-enhance-event-event-temporal-relation-extraction</pwccode>
     </paper>
     <paper id="816">
       <title>Separating Retention from Extraction in the Evaluation of End-to-end <fixed-case>R</fixed-case>elation <fixed-case>E</fixed-case>xtraction</title>
@@ -12684,8 +12692,8 @@
       <url hash="5e46e050">2021.emnlp-main.816</url>
       <bibkey>taille-etal-2021-separating</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.816</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
       <video href="2021.emnlp-main.816.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
     </paper>
     <paper id="817">
       <title>Automatic Text Evaluation through the Lens of <fixed-case>W</fixed-case>asserstein Barycenters</title>
@@ -12698,10 +12706,10 @@
       <url hash="db428beb">2021.emnlp-main.817</url>
       <bibkey>colombo-etal-2021-automatic</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.817</doi>
+      <video href="2021.emnlp-main.817.mp4"/>
       <pwccode url="" additional="true"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2016">WMT 2016</pwcdataset>
-      <video href="2021.emnlp-main.817.mp4"/>
     </paper>
     <paper id="818">
       <title>Visually Grounded Reasoning across Languages and Cultures</title>
@@ -12716,11 +12724,11 @@
       <url hash="60cbe204">2021.emnlp-main.818</url>
       <bibkey>liu-etal-2021-visually</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.818</doi>
+      <video href="2021.emnlp-main.818.mp4"/>
       <pwccode url="https://github.com/e-bug/volta" additional="true">e-bug/volta</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/marvl">MaRVL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iglue">IGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/nlvr">NLVR</pwcdataset>
-      <video href="2021.emnlp-main.818.mp4"/>
     </paper>
     <paper id="819">
       <title>Back to Square One: Artifact Detection, Training and Commonsense Disentanglement in the <fixed-case>W</fixed-case>inograd Schema</title>
@@ -12733,9 +12741,9 @@
       <url hash="72210213">2021.emnlp-main.819</url>
       <bibkey>elazar-etal-2021-back</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.819</doi>
+      <video href="2021.emnlp-main.819.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winogrande">WinoGrande</pwcdataset>
-      <video href="2021.emnlp-main.819.mp4"/>
     </paper>
     <paper id="820">
       <title>Robustness Evaluation of Entity Disambiguation Using Prior Probes: the Case of Entity Overshadowing</title>
@@ -12762,13 +12770,13 @@
       <url hash="19c57970">2021.emnlp-main.821</url>
       <bibkey>mahendra-etal-2021-indonli</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.821</doi>
+      <video href="2021.emnlp-main.821.mp4"/>
       <pwccode url="https://github.com/ir-nlp-csui/indonli" additional="false">ir-nlp-csui/indonli</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/indonli">IndoNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/indonlu-benchmark">IndoNLU Benchmark</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ocnli">OCNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.emnlp-main.821.mp4"/>
     </paper>
     <paper id="822">
       <title>Agreeing to Disagree: Annotating Offensive Language Datasets with Annotators’ Disagreement</title>
@@ -12793,8 +12801,8 @@
       <url hash="75b9148b">2021.emnlp-main.823</url>
       <bibkey>stanojevic-cohen-2021-root</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.823</doi>
-      <pwccode url="https://github.com/stanojevic/fast-mst-algorithm" additional="false">stanojevic/fast-mst-algorithm</pwccode>
       <video href="2021.emnlp-main.823.mp4"/>
+      <pwccode url="https://github.com/stanojevic/fast-mst-algorithm" additional="false">stanojevic/fast-mst-algorithm</pwccode>
     </paper>
     <paper id="824">
       <title>Efficient Sampling of Dependency Structure</title>
@@ -12806,8 +12814,8 @@
       <url hash="2c5b6dd2">2021.emnlp-main.824</url>
       <bibkey>zmigrod-etal-2021-efficient</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.824</doi>
-      <pwccode url="https://github.com/rycolab/treesample" additional="false">rycolab/treesample</pwccode>
       <video href="2021.emnlp-main.824.mp4"/>
+      <pwccode url="https://github.com/rycolab/treesample" additional="false">rycolab/treesample</pwccode>
     </paper>
     <paper id="825">
       <title>Reducing Discontinuous to Continuous Parsing with Pointer Network Reordering</title>
@@ -12818,9 +12826,9 @@
       <url hash="dd393819">2021.emnlp-main.825</url>
       <bibkey>fernandez-gonzalez-gomez-rodriguez-2021-reducing</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.825</doi>
+      <video href="2021.emnlp-main.825.mp4"/>
       <pwccode url="https://github.com/danifg/Pointer-Network-Reordering" additional="false">danifg/Pointer-Network-Reordering</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.emnlp-main.825.mp4"/>
     </paper>
     <paper id="826">
       <title>A New Representation for Span-based <fixed-case>CCG</fixed-case> Parsing</title>
@@ -12831,8 +12839,8 @@
       <url hash="fc23b410">2021.emnlp-main.826</url>
       <bibkey>kato-matsubara-2021-new</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.826</doi>
-      <pwccode url="https://github.com/yosihide/span-based-ccg-derivation" additional="false">yosihide/span-based-ccg-derivation</pwccode>
       <video href="2021.emnlp-main.826.mp4"/>
+      <pwccode url="https://github.com/yosihide/span-based-ccg-derivation" additional="false">yosihide/span-based-ccg-derivation</pwccode>
     </paper>
     <paper id="827">
       <title><fixed-case>W</fixed-case>hat to Pre-Train on? <fixed-case>E</fixed-case>fficient Intermediate Task Selection</title>
@@ -12845,6 +12853,7 @@
       <url hash="15ba0e3a">2021.emnlp-main.827</url>
       <bibkey>poth-etal-2021-pre</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.827</doi>
+      <video href="2021.emnlp-main.827.mp4"/>
       <pwccode url="https://github.com/adapter-hub/efficient-task-transfer" additional="false">adapter-hub/efficient-task-transfer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
@@ -12852,7 +12861,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.emnlp-main.827.mp4"/>
     </paper>
     <paper id="828">
       <title><fixed-case>P</fixed-case>ermute<fixed-case>F</fixed-case>ormer: Efficient Relative Position Encoding for Long Sequences</title>
@@ -12862,9 +12870,10 @@
       <url hash="04b30d1c">2021.emnlp-main.828</url>
       <bibkey>chen-2021-permuteformer</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.828</doi>
+      <video href="2021.emnlp-main.828.mp4"/>
       <pwccode url="https://github.com/cpcp1998/permuteformer" additional="false">cpcp1998/permuteformer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.emnlp-main.828.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="829">
       <title>Block Pruning For Faster Transformers</title>
@@ -12877,13 +12886,13 @@
       <url hash="38185732">2021.emnlp-main.829</url>
       <bibkey>lagunas-etal-2021-block</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.829</doi>
+      <video href="2021.emnlp-main.829.mp4"/>
       <pwccode url="https://github.com/huggingface/nn_pruning" additional="false">huggingface/nn_pruning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quora-question-pairs">Quora Question Pairs</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.829.mp4"/>
     </paper>
     <paper id="830">
       <title>Finetuning Pretrained Transformers into <fixed-case>RNN</fixed-case>s</title>
@@ -12901,9 +12910,10 @@
       <url hash="20e04c33">2021.emnlp-main.830</url>
       <bibkey>kasai-etal-2021-finetuning</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.830</doi>
+      <video href="2021.emnlp-main.830.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.emnlp-main.830.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="831">
       <title>How to Train <fixed-case>BERT</fixed-case> with an Academic Budget</title>
@@ -12915,7 +12925,8 @@
       <url hash="af151510">2021.emnlp-main.831</url>
       <bibkey>izsak-etal-2021-train</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.831</doi>
-      <pwccode url="https://github.com/peteriz/academic-budget-bert" additional="true">peteriz/academic-budget-bert</pwccode>
+      <video href="2021.emnlp-main.831.mp4"/>
+      <pwccode url="https://github.com/IntelLabs/academic-budget-bert" additional="true">IntelLabs/academic-budget-bert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
@@ -12926,7 +12937,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sts-benchmark">STS Benchmark</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.emnlp-main.831.mp4"/>
     </paper>
     <paper id="832">
       <title>Beyond Preserved Accuracy: Evaluating Loyalty and Robustness of <fixed-case>BERT</fixed-case> Compression</title>
@@ -12941,9 +12951,9 @@
       <url hash="c297e6dc">2021.emnlp-main.832</url>
       <bibkey>xu-etal-2021-beyond</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.832</doi>
+      <video href="2021.emnlp-main.832.mp4"/>
       <pwccode url="https://github.com/jetrunner/beyond-preserved-accuracy" additional="false">jetrunner/beyond-preserved-accuracy</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
-      <video href="2021.emnlp-main.832.mp4"/>
     </paper>
     <paper id="833">
       <title><fixed-case>I</fixed-case>ndo<fixed-case>BERT</fixed-case>weet: A Pretrained Language Model for <fixed-case>I</fixed-case>ndonesian <fixed-case>T</fixed-case>witter with Effective Domain-Specific Vocabulary Initialization</title>
@@ -12955,8 +12965,8 @@
       <url hash="6045b071">2021.emnlp-main.833</url>
       <bibkey>koto-etal-2021-indobertweet</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.833</doi>
-      <pwccode url="https://github.com/indolem/indobertweet" additional="false">indolem/indobertweet</pwccode>
       <video href="2021.emnlp-main.833.mp4"/>
+      <pwccode url="https://github.com/indolem/indobertweet" additional="false">indolem/indobertweet</pwccode>
     </paper>
     <paper id="834">
       <title>Pushing on Text Readability Assessment: A Transformer Meets Handcrafted Linguistic Features</title>
@@ -12970,9 +12980,9 @@
       <revision id="1" href="2021.emnlp-main.834v1" hash="68ff98b2"/>
       <revision id="2" href="2021.emnlp-main.834v2" hash="75ab197c" date="2021-11-09">Corrected a typo.</revision>
       <doi>10.18653/v1/2021.emnlp-main.834</doi>
+      <video href="2021.emnlp-main.834.mp4"/>
       <pwccode url="https://github.com/brucewlee/lingfeat" additional="false">brucewlee/lingfeat</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/onestopenglish">OneStopEnglish</pwcdataset>
-      <video href="2021.emnlp-main.834.mp4"/>
     </paper>
     <paper id="835">
       <title>Types of Out-of-Distribution Texts and How to Detect Them</title>
@@ -12984,6 +12994,7 @@
       <url hash="098c434c">2021.emnlp-main.835</url>
       <bibkey>arora-etal-2021-types</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.835</doi>
+      <video href="2021.emnlp-main.835.mp4"/>
       <pwccode url="https://github.com/uditarora/ood-text-emnlp" additional="false">uditarora/ood-text-emnlp</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dbpedia">DBpedia</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
@@ -12992,7 +13003,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/rte">RTE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.emnlp-main.835.mp4"/>
     </paper>
     <paper id="836">
       <title>Self-training with Few-shot Rationalization</title>
@@ -13004,10 +13014,10 @@
       <url hash="4c038751">2021.emnlp-main.836</url>
       <bibkey>bhat-etal-2021-self</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.836</doi>
+      <video href="2021.emnlp-main.836.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/e-snli">e-SNLI</pwcdataset>
-      <video href="2021.emnlp-main.836.mp4"/>
     </paper>
     <paper id="837">
       <title><fixed-case>MTA</fixed-case>dam: Automatic Balancing of Multiple Training Loss Terms</title>
@@ -13018,10 +13028,10 @@
       <url hash="1b91b669">2021.emnlp-main.837</url>
       <bibkey>malkiel-wolf-2021-mtadam</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.837</doi>
+      <video href="2021.emnlp-main.837.mp4"/>
       <pwccode url="https://github.com/ItzikMalkiel/MTAdam" additional="false">ItzikMalkiel/MTAdam</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bsd">BSD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/set14">Set14</pwcdataset>
-      <video href="2021.emnlp-main.837.mp4"/>
     </paper>
     <paper id="838">
       <title>Softmax Tree: An Accurate, Fast Classifier When the Number of Classes Is Large</title>
@@ -13077,8 +13087,8 @@
       <attachment type="Software" hash="e2591cf4">2021.emnlp-main.841.Software.zip</attachment>
       <bibkey>basu-roy-chowdhury-etal-2021-everything</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.841</doi>
-      <pwccode url="https://github.com/fabrahman/rebart" additional="false">fabrahman/rebart</pwccode>
       <video href="2021.emnlp-main.841.mp4"/>
+      <pwccode url="https://github.com/fabrahman/rebart" additional="false">fabrahman/rebart</pwccode>
     </paper>
     <paper id="842">
       <title><fixed-case>V</fixed-case>ee<fixed-case>A</fixed-case>lign: Multifaceted Context Representation Using Dual Attention for Ontology Alignment</title>
@@ -13091,8 +13101,8 @@
       <attachment type="Software" hash="cb52034e">2021.emnlp-main.842.Software.zip</attachment>
       <bibkey>iyer-etal-2021-veealign</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.842</doi>
-      <pwccode url="https://github.com/remorax/veealign" additional="false">remorax/veealign</pwccode>
       <video href="2021.emnlp-main.842.mp4"/>
+      <pwccode url="https://github.com/remorax/veealign" additional="false">remorax/veealign</pwccode>
     </paper>
     <paper id="843">
       <title>Finding needles in a haystack: Sampling Structurally-diverse Training Sets from Synthetic Data for Compositional Generalization</title>
@@ -13104,8 +13114,8 @@
       <url hash="a0fa7271">2021.emnlp-main.843</url>
       <bibkey>oren-etal-2021-finding</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.843</doi>
-      <pwccode url="https://github.com/inbaroren/scfg-sampling-for-comp-gen" additional="false">inbaroren/scfg-sampling-for-comp-gen</pwccode>
       <video href="2021.emnlp-main.843.mp4"/>
+      <pwccode url="https://github.com/inbaroren/scfg-sampling-for-comp-gen" additional="false">inbaroren/scfg-sampling-for-comp-gen</pwccode>
     </paper>
     <paper id="844">
       <title><fixed-case>G</fixed-case>ene<fixed-case>S</fixed-case>is: <fixed-case>A</fixed-case> <fixed-case>G</fixed-case>enerative <fixed-case>A</fixed-case>pproach to <fixed-case>S</fixed-case>ubstitutes in <fixed-case>C</fixed-case>ontext</title>
@@ -13117,8 +13127,8 @@
       <url hash="f3fb91e0">2021.emnlp-main.844</url>
       <bibkey>lacerra-etal-2021-genesis</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.844</doi>
-      <pwccode url="https://github.com/sapienzanlp/genesis" additional="false">sapienzanlp/genesis</pwccode>
       <video href="2021.emnlp-main.844.mp4"/>
+      <pwccode url="https://github.com/sapienzanlp/genesis" additional="false">sapienzanlp/genesis</pwccode>
     </paper>
     <paper id="845">
       <title>Semi-Supervised Exaggeration Detection of Health Science Press Releases</title>
@@ -13129,8 +13139,8 @@
       <url hash="7e0fa396">2021.emnlp-main.845</url>
       <bibkey>wright-augenstein-2021-semi</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.845</doi>
-      <pwccode url="https://github.com/copenlu/scientific-exaggeration-detection" additional="false">copenlu/scientific-exaggeration-detection</pwccode>
       <video href="2021.emnlp-main.845.mp4"/>
+      <pwccode url="https://github.com/copenlu/scientific-exaggeration-detection" additional="false">copenlu/scientific-exaggeration-detection</pwccode>
     </paper>
     <paper id="846">
       <title>Phrase-<fixed-case>BERT</fixed-case>: Improved Phrase Embeddings from <fixed-case>BERT</fixed-case> with an Application to Corpus Exploration</title>
@@ -13142,11 +13152,11 @@
       <url hash="cbf3d88a">2021.emnlp-main.846</url>
       <bibkey>wang-etal-2021-phrase</bibkey>
       <doi>10.18653/v1/2021.emnlp-main.846</doi>
+      <video href="2021.emnlp-main.846.mp4"/>
       <pwccode url="https://github.com/sf-wa-326/phrase-bert-topic-model" additional="false">sf-wa-326/phrase-bert-topic-model</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/big-bird">BiRD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws">PAWS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/the-pile">The Pile</pwcdataset>
-      <video href="2021.emnlp-main.846.mp4"/>
     </paper>
     <paper id="847">
       <title>Detecting Contact-Induced Semantic Shifts: <fixed-case>W</fixed-case>hat Can Embedding-Based Methods Do in Practice?</title>
@@ -13223,8 +13233,8 @@
       <url hash="e85bc09f">2021.emnlp-demo.4</url>
       <bibkey>steffen-van-genabith-2021-transins</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.4</doi>
-      <pwccode url="https://github.com/dfki-mlt/transins" additional="false">dfki-mlt/transins</pwccode>
       <video href="2021.emnlp-demo.4.mp4"/>
+      <pwccode url="https://github.com/dfki-mlt/transins" additional="false">dfki-mlt/transins</pwccode>
     </paper>
     <paper id="5">
       <title><fixed-case>ET</fixed-case>: A Workstation for Querying, Editing and Evaluating Annotated Corpora</title>
@@ -13235,8 +13245,8 @@
       <url hash="07ec1285">2021.emnlp-demo.5</url>
       <bibkey>de-souza-freitas-2021-et</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.5</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
       <video href="2021.emnlp-demo.5.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
     </paper>
     <paper id="6">
       <title>N-<fixed-case>LTP</fixed-case>: An Open-source Neural Language Technology Platform for <fixed-case>C</fixed-case>hinese</title>
@@ -13249,8 +13259,8 @@
       <url hash="76117ec7">2021.emnlp-demo.6</url>
       <bibkey>che-etal-2021-n</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.6</doi>
-      <pwccode url="https://github.com/HIT-SCIR/ltp" additional="false">HIT-SCIR/ltp</pwccode>
       <video href="2021.emnlp-demo.6.mp4"/>
+      <pwccode url="https://github.com/HIT-SCIR/ltp" additional="false">HIT-SCIR/ltp</pwccode>
     </paper>
     <paper id="7">
       <title><fixed-case>COMBO</fixed-case>: State-of-the-Art Morphosyntactic Analysis</title>
@@ -13319,12 +13329,12 @@
       <attachment type="Software" hash="1983c5a4">2021.emnlp-demo.11.Software.zip</attachment>
       <bibkey>feldhus-etal-2021-thermostat</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.11</doi>
+      <video href="2021.emnlp-demo.11.mp4"/>
       <pwccode url="https://github.com/dfki-nlp/thermostat" additional="true">dfki-nlp/thermostat</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.emnlp-demo.11.mp4"/>
     </paper>
     <paper id="12">
       <title><fixed-case>LM</fixed-case>diff: A Visual Diff Tool to Compare Language Models</title>
@@ -13337,10 +13347,10 @@
       <url hash="2bac39a1">2021.emnlp-demo.12</url>
       <bibkey>strobelt-etal-2021-lmdiff</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.12</doi>
+      <video href="2021.emnlp-demo.12.mp4"/>
       <pwccode url="https://github.com/hendrikstrobelt/lmdiff" additional="false">hendrikstrobelt/lmdiff</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/commonsenseqa">CommonsenseQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winobias">WinoBias</pwcdataset>
-      <video href="2021.emnlp-demo.12.mp4"/>
     </paper>
     <paper id="13">
       <title>Semantic Context Path Labeling for Semantic Exploration of User Reviews</title>
@@ -13365,13 +13375,13 @@
       <url hash="d19ba3fb">2021.emnlp-demo.14</url>
       <bibkey>vath-etal-2021-beyond</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.14</doi>
+      <video href="2021.emnlp-demo.14.mp4"/>
       <pwccode url="https://github.com/patilli/vqa_benchmarking" additional="false">patilli/vqa_benchmarking</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clevr">CLEVR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gqa">GQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ok-vqa">OK-VQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/textvqa">TextVQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.emnlp-demo.14.mp4"/>
     </paper>
     <paper id="15">
       <title>Athena 2.0: Contextualized Dialogue Management for an <fixed-case>A</fixed-case>lexa <fixed-case>P</fixed-case>rize <fixed-case>S</fixed-case>ocial<fixed-case>B</fixed-case>ot</title>
@@ -13409,8 +13419,8 @@
       <url hash="086e7258">2021.emnlp-demo.16</url>
       <bibkey>blloshmi-etal-2021-spring</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.16</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/bio">Bio</pwcdataset>
       <video href="2021.emnlp-demo.16.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/bio">Bio</pwcdataset>
     </paper>
     <paper id="17">
       <title>fairseq Sˆ2: A Scalable and Integrable Speech Synthesis Toolkit</title>
@@ -13509,11 +13519,11 @@
       <url hash="e870681f">2021.emnlp-demo.21</url>
       <bibkey>lhoest-etal-2021-datasets</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.21</doi>
+      <video href="2021.emnlp-demo.21.mp4"/>
       <pwccode url="https://github.com/huggingface/datasets" additional="false">huggingface/datasets</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.emnlp-demo.21.mp4"/>
     </paper>
     <paper id="22">
       <title>Summary Explorer: Visualizing the State of the Art in Text Summarization</title>
@@ -13527,8 +13537,8 @@
       <url hash="f2634d04">2021.emnlp-demo.22</url>
       <bibkey>syed-etal-2021-summary</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.22</doi>
-      <pwccode url="https://github.com/webis-de/summary-explorer" additional="false">webis-de/summary-explorer</pwccode>
       <video href="2021.emnlp-demo.22.mp4"/>
+      <pwccode url="https://github.com/webis-de/summary-explorer" additional="false">webis-de/summary-explorer</pwccode>
     </paper>
     <paper id="23">
       <title><fixed-case>M</fixed-case>eet<fixed-case>D</fixed-case>ot: Videoconferencing with Live Translation Captions</title>
@@ -13561,8 +13571,8 @@
       <url hash="2eaeb3bf">2021.emnlp-demo.24</url>
       <bibkey>chheda-etal-2021-box</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.24</doi>
-      <pwccode url="https://github.com/iesl/box-embeddings" additional="false">iesl/box-embeddings</pwccode>
       <video href="2021.emnlp-demo.24.mp4"/>
+      <pwccode url="https://github.com/iesl/box-embeddings" additional="false">iesl/box-embeddings</pwccode>
     </paper>
     <paper id="25">
       <title><fixed-case>L</fixed-case>exi<fixed-case>C</fixed-case>lean: An annotation tool for rapid multi-task lexical normalisation</title>
@@ -13576,8 +13586,8 @@
       <url hash="d0a345c6">2021.emnlp-demo.25</url>
       <bibkey>bikaun-etal-2021-lexiclean</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.25</doi>
-      <pwccode url="https://github.com/nlp-tlp/lexiclean" additional="false">nlp-tlp/lexiclean</pwccode>
       <video href="2021.emnlp-demo.25.mp4"/>
+      <pwccode url="https://github.com/nlp-tlp/lexiclean" additional="false">nlp-tlp/lexiclean</pwccode>
     </paper>
     <paper id="26">
       <title>T3-Vis: visual analytic for Training and fine-Tuning Transformers in <fixed-case>NLP</fixed-case></title>
@@ -13591,8 +13601,8 @@
       <url hash="acef0e4c">2021.emnlp-demo.26</url>
       <bibkey>li-etal-2021-t3</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.26</doi>
-      <pwccode url="https://github.com/raymondzmc/t3-vis" additional="false">raymondzmc/t3-vis</pwccode>
       <video href="2021.emnlp-demo.26.mp4"/>
+      <pwccode url="https://github.com/raymondzmc/t3-vis" additional="false">raymondzmc/t3-vis</pwccode>
     </paper>
     <paper id="27">
       <title><fixed-case>D</fixed-case>omi<fixed-case>K</fixed-case>now<fixed-case>S</fixed-case>: A Library for Integration of Symbolic Domain Knowledge in Deep Learning</title>
@@ -13606,8 +13616,8 @@
       <url hash="aea8384f">2021.emnlp-demo.27</url>
       <bibkey>rajaby-faghihi-etal-2021-domiknows</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.27</doi>
-      <pwccode url="https://github.com/hlr/domiknows" additional="false">hlr/domiknows</pwccode>
       <video href="2021.emnlp-demo.27.mp4"/>
+      <pwccode url="https://github.com/hlr/domiknows" additional="false">hlr/domiknows</pwccode>
     </paper>
     <paper id="28">
       <title><fixed-case>O</fixed-case>pen<fixed-case>F</fixed-case>raming: Open-sourced Tool for Computational Framing Analysis of Multilingual Data</title>
@@ -13628,8 +13638,8 @@
       <url hash="42c2be30">2021.emnlp-demo.28</url>
       <bibkey>bhatia-etal-2021-openframing</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.28</doi>
-      <pwccode url="https://github.com/vibss2397/openframing" additional="false">vibss2397/openframing</pwccode>
       <video href="2021.emnlp-demo.28.mp4"/>
+      <pwccode url="https://github.com/vibss2397/openframing" additional="false">vibss2397/openframing</pwccode>
     </paper>
     <paper id="29">
       <title><fixed-case>I</fixed-case>r<fixed-case>E</fixed-case>ne-viz: Visualizing Energy Consumption of Transformer Models</title>
@@ -13657,10 +13667,10 @@
       <url hash="2ca70ecf">2021.emnlp-demo.30</url>
       <bibkey>levy-etal-2021-open</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.30</doi>
+      <video href="2021.emnlp-demo.30.mp4"/>
       <pwccode url="https://github.com/sharonlevy/open_domain_covidqa" additional="false">sharonlevy/open_domain_covidqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cord-19">CORD-19</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/covidqa">CovidQA</pwcdataset>
-      <video href="2021.emnlp-demo.30.mp4"/>
     </paper>
     <paper id="31">
       <title>Project <fixed-case>D</fixed-case>ebater <fixed-case>API</fixed-case>s: <fixed-case>D</fixed-case>ecomposing the <fixed-case>AI</fixed-case> Grand Challenge</title>
@@ -13674,8 +13684,8 @@
       <url hash="51d3d1b2">2021.emnlp-demo.31</url>
       <bibkey>bar-haim-etal-2021-project</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.31</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/ibm-rank-30k">IBM-Rank-30k</pwcdataset>
       <video href="2021.emnlp-demo.31.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/ibm-rank-30k">IBM-Rank-30k</pwcdataset>
     </paper>
     <paper id="32">
       <title><fixed-case>C</fixed-case>ro<fixed-case>A</fixed-case>no : A Crowd Annotation Platform for Improving Label Consistency of <fixed-case>C</fixed-case>hinese <fixed-case>NER</fixed-case> Dataset</title>
@@ -13712,10 +13722,10 @@
       <url hash="46621a34">2021.emnlp-demo.33</url>
       <bibkey>hirsch-etal-2021-ifacetsum</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.33</doi>
+      <video href="2021.emnlp-demo.33.mp4"/>
       <pwccode url="https://github.com/biu-nlp/ifacetsum" additional="false">biu-nlp/ifacetsum</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ecb">ECB+</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wec-eng">WEC-Eng</pwcdataset>
-      <video href="2021.emnlp-demo.33.mp4"/>
     </paper>
     <paper id="34">
       <title><fixed-case>AMuSE-WSD</fixed-case>: <fixed-case>A</fixed-case>n All-in-one Multilingual System for Easy <fixed-case>W</fixed-case>ord <fixed-case>S</fixed-case>ense <fixed-case>D</fixed-case>isambiguation</title>
@@ -13729,8 +13739,8 @@
       <url hash="b66d4835">2021.emnlp-demo.34</url>
       <bibkey>orlando-etal-2021-amuse</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.34</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
       <video href="2021.emnlp-demo.34.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
     </paper>
     <paper id="35">
       <title><fixed-case>S</fixed-case>eq<fixed-case>A</fixed-case>ttack: <fixed-case>O</fixed-case>n Adversarial Attacks for Named Entity Recognition</title>
@@ -13741,8 +13751,8 @@
       <url hash="874d5009">2021.emnlp-demo.35</url>
       <bibkey>simoncini-spanakis-2021-seqattack</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.35</doi>
-      <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <video href="2021.emnlp-demo.35.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
     </paper>
     <paper id="36">
       <title><fixed-case>InVeRo-XL</fixed-case>: <fixed-case>M</fixed-case>aking Cross-Lingual <fixed-case>S</fixed-case>emantic <fixed-case>R</fixed-case>ole <fixed-case>L</fixed-case>abeling Accessible with Intelligible Verbs and Roles</title>
@@ -13773,6 +13783,7 @@
       <url hash="46160b2b">2021.emnlp-demo.37</url>
       <bibkey>ni-etal-2021-summertime</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.37</doi>
+      <video href="2021.emnlp-demo.37.mp4"/>
       <pwccode url="https://github.com/Yale-LILY/SummerTime" additional="false">Yale-LILY/SummerTime</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mlsum">MLSUM</pwcdataset>
@@ -13780,7 +13791,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/pubmedqa">PubMedQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scisummnet">ScisummNet</pwcdataset>
-      <video href="2021.emnlp-demo.37.mp4"/>
     </paper>
     <paper id="38">
       <title>Chandler: An Explainable Sarcastic Response Generator</title>
@@ -13805,8 +13815,8 @@
       <url hash="ccbfb956">2021.emnlp-demo.39</url>
       <bibkey>jain-etal-2021-tabpert</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.39</doi>
-      <pwccode url="https://github.com/utahnlp/tabpert" additional="false">utahnlp/tabpert</pwccode>
       <video href="2021.emnlp-demo.39.mp4"/>
+      <pwccode url="https://github.com/utahnlp/tabpert" additional="false">utahnlp/tabpert</pwccode>
     </paper>
     <paper id="40">
       <title><fixed-case>DRIFT</fixed-case>: A Toolkit for Diachronic Analysis of Scientific Literature</title>
@@ -13820,8 +13830,8 @@
       <attachment type="Software" hash="1aa6ffe5">2021.emnlp-demo.40.Software.zip</attachment>
       <bibkey>sharma-etal-2021-drift</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.40</doi>
-      <pwccode url="https://github.com/rajaswa/DRIFT" additional="false">rajaswa/DRIFT</pwccode>
       <video href="2021.emnlp-demo.40.mp4"/>
+      <pwccode url="https://github.com/rajaswa/DRIFT" additional="false">rajaswa/DRIFT</pwccode>
     </paper>
     <paper id="41">
       <title><fixed-case>FAST</fixed-case>: <fixed-case>F</fixed-case>ast <fixed-case>A</fixed-case>nnotation tool for <fixed-case>S</fixed-case>mar<fixed-case>T</fixed-case> devices</title>
@@ -13835,8 +13845,8 @@
       <attachment type="Software" hash="d6af4d7e">2021.emnlp-demo.41.Software.zip</attachment>
       <bibkey>kawamoto-etal-2021-fast</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.41</doi>
-      <pwccode url="https://github.com/cyberagent/fast-annotation-tool" additional="false">cyberagent/fast-annotation-tool</pwccode>
       <video href="2021.emnlp-demo.41.mp4"/>
+      <pwccode url="https://github.com/cyberagent/fast-annotation-tool" additional="false">cyberagent/fast-annotation-tool</pwccode>
     </paper>
     <paper id="42">
       <title>deep<fixed-case>Q</fixed-case>uest-py: <fixed-case>L</fixed-case>arge and Distilled Models for Quality Estimation</title>
@@ -13851,9 +13861,9 @@
       <url hash="71248731">2021.emnlp-demo.42</url>
       <bibkey>alva-manchego-etal-2021-deepquest</bibkey>
       <doi>10.18653/v1/2021.emnlp-demo.42</doi>
+      <video href="2021.emnlp-demo.42.mp4"/>
       <pwccode url="https://github.com/sheffieldnlp/deepQuest-py" additional="false">sheffieldnlp/deepQuest-py</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqe">MLQE</pwcdataset>
-      <video href="2021.emnlp-demo.42.mp4"/>
     </paper>
   </volume>
   <volume id="tutorials" ingest-date="2021-11-04">
diff --git a/data/xml/2021.findings.xml b/data/xml/2021.findings.xml
index c89843f2f7..e3d2675c89 100644
--- a/data/xml/2021.findings.xml
+++ b/data/xml/2021.findings.xml
@@ -4942,6 +4942,7 @@
       <pwcdataset url="https://paperswithcode.com/dataset/coin">COIN</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/crosstask">CrossTask</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/howto100m">HowTo100M</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/msr-vtt">MSR-VTT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/youcook2">YouCook2</pwcdataset>
     </paper>
     <paper id="371">
diff --git a/data/xml/2021.naacl.xml b/data/xml/2021.naacl.xml
index 0995817c65..06082dd161 100644
--- a/data/xml/2021.naacl.xml
+++ b/data/xml/2021.naacl.xml
@@ -68,8 +68,8 @@
       <url hash="bc4f4426">2021.naacl-main.4</url>
       <doi>10.18653/v1/2021.naacl-main.4</doi>
       <bibkey>zhang-ji-2021-abstract</bibkey>
-      <pwccode url="https://github.com/zhangzx-uiuc/amr-ie" additional="false">zhangzx-uiuc/amr-ie</pwccode>
       <video href="2021.naacl-main.4.mp4"/>
+      <pwccode url="https://github.com/zhangzx-uiuc/amr-ie" additional="false">zhangzx-uiuc/amr-ie</pwccode>
     </paper>
     <paper id="5">
       <title>A Frustratingly Easy Approach for Entity and Relation Extraction</title>
@@ -80,11 +80,11 @@
       <url hash="830996dd">2021.naacl-main.5</url>
       <doi>10.18653/v1/2021.naacl-main.5</doi>
       <bibkey>zhong-chen-2021-frustratingly</bibkey>
+      <video href="2021.naacl-main.5.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/PURE" additional="true">princeton-nlp/PURE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2004">ACE 2004</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ace-2005">ACE 2005</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
-      <video href="2021.naacl-main.5.mp4"/>
     </paper>
     <paper id="6">
       <title>Event Time Extraction and Propagation via Graph Attention Networks</title>
@@ -101,8 +101,8 @@
       <url hash="f285cc13">2021.naacl-main.6</url>
       <doi>10.18653/v1/2021.naacl-main.6</doi>
       <bibkey>wen-etal-2021-event</bibkey>
-      <pwccode url="https://github.com/wenhycs/naacl2021-event-time-extraction-and-propagation-via-graph-attention-networks" additional="false">wenhycs/naacl2021-event-time-extraction-and-propagation-via-graph-attention-networks</pwccode>
       <video href="2021.naacl-main.6.mp4"/>
+      <pwccode url="https://github.com/wenhycs/naacl2021-event-time-extraction-and-propagation-via-graph-attention-networks" additional="false">wenhycs/naacl2021-event-time-extraction-and-propagation-via-graph-attention-networks</pwccode>
     </paper>
     <paper id="7">
       <title>Probing Word Translations in the Transformer and Trading Decoder for Encoder Layers</title>
@@ -127,8 +127,8 @@
       <url hash="b2235f73">2021.naacl-main.8</url>
       <doi>10.18653/v1/2021.naacl-main.8</doi>
       <bibkey>slobodkin-etal-2021-mediators</bibkey>
-      <pwccode url="https://github.com/lovodkin93/BERT-context-distance" additional="false">lovodkin93/BERT-context-distance</pwccode>
       <video href="2021.naacl-main.8.mp4"/>
+      <pwccode url="https://github.com/lovodkin93/BERT-context-distance" additional="false">lovodkin93/BERT-context-distance</pwccode>
     </paper>
     <paper id="9">
       <title>Automatic Generation of Contrast Sets from Scene Graphs: Probing the Compositional Consistency of <fixed-case>GQA</fixed-case></title>
@@ -141,9 +141,9 @@
       <url hash="1236a480">2021.naacl-main.9</url>
       <doi>10.18653/v1/2021.naacl-main.9</doi>
       <bibkey>bitton-etal-2021-automatic</bibkey>
+      <video href="2021.naacl-main.9.mp4"/>
       <pwccode url="https://github.com/yonatanbitton/AutoGenOfContrastSetsFromSceneGraphs" additional="true">yonatanbitton/AutoGenOfContrastSetsFromSceneGraphs</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gqa">GQA</pwcdataset>
-      <video href="2021.naacl-main.9.mp4"/>
     </paper>
     <paper id="10">
       <title>Multilingual Language Models Predict Human Reading Behavior</title>
@@ -157,8 +157,8 @@
       <url hash="c0b8e99c">2021.naacl-main.10</url>
       <doi>10.18653/v1/2021.naacl-main.10</doi>
       <bibkey>hollenstein-etal-2021-multilingual</bibkey>
-      <pwccode url="https://github.com/DS3Lab/multilingual-gaze" additional="false">DS3Lab/multilingual-gaze</pwccode>
       <video href="2021.naacl-main.10.mp4"/>
+      <pwccode url="https://github.com/DS3Lab/multilingual-gaze" additional="false">DS3Lab/multilingual-gaze</pwccode>
     </paper>
     <paper id="11">
       <title>Do Syntactic Probes Probe Syntax? Experiments with Jabberwocky Probing</title>
@@ -169,8 +169,8 @@
       <url hash="4029576a">2021.naacl-main.11</url>
       <doi>10.18653/v1/2021.naacl-main.11</doi>
       <bibkey>hall-maudslay-cotterell-2021-syntactic</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
       <video href="2021.naacl-main.11.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
     </paper>
     <paper id="12">
       <title>A Non-Linear Structural Probe</title>
@@ -196,8 +196,8 @@
       <url hash="77b4727f">2021.naacl-main.13</url>
       <doi>10.18653/v1/2021.naacl-main.13</doi>
       <bibkey>wallace-etal-2021-concealed</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <video href="2021.naacl-main.13.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
     </paper>
     <paper id="14">
       <title>Backtranslation Feedback Improves User Confidence in <fixed-case>MT</fixed-case>, Not Quality</title>
@@ -217,8 +217,8 @@
       <attachment type="OptionalSupplementaryCode" hash="4c946467">2021.naacl-main.14.OptionalSupplementaryCode.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.14</doi>
       <bibkey>zouhar-etal-2021-backtranslation</bibkey>
-      <pwccode url="https://github.com/zouharvi/ptakopet" additional="false">zouharvi/ptakopet</pwccode>
       <video href="2021.naacl-main.14.mp4"/>
+      <pwccode url="https://github.com/zouharvi/ptakopet" additional="false">zouharvi/ptakopet</pwccode>
     </paper>
     <paper id="15">
       <title>Data Filtering using Cross-Lingual Word Embeddings</title>
@@ -243,8 +243,8 @@
       <url hash="71a775df">2021.naacl-main.16</url>
       <doi>10.18653/v1/2021.naacl-main.16</doi>
       <bibkey>chronopoulou-etal-2021-improving</bibkey>
-      <pwccode url="https://github.com/alexandra-chron/lexical_xlm_relm" additional="false">alexandra-chron/lexical_xlm_relm</pwccode>
       <video href="2021.naacl-main.16.mp4"/>
+      <pwccode url="https://github.com/alexandra-chron/lexical_xlm_relm" additional="false">alexandra-chron/lexical_xlm_relm</pwccode>
     </paper>
     <paper id="17">
       <title>Neural Machine Translation without Embeddings</title>
@@ -255,8 +255,8 @@
       <url hash="312cdd66">2021.naacl-main.17</url>
       <doi>10.18653/v1/2021.naacl-main.17</doi>
       <bibkey>shaham-levy-2021-neural</bibkey>
-      <pwccode url="https://github.com/UriSha/EmbeddinglessNMT" additional="true">UriSha/EmbeddinglessNMT</pwccode>
       <video href="2021.naacl-main.17.mp4"/>
+      <pwccode url="https://github.com/UriSha/EmbeddinglessNMT" additional="true">UriSha/EmbeddinglessNMT</pwccode>
     </paper>
     <paper id="18">
       <title>Counterfactual Data Augmentation for Neural Machine Translation</title>
@@ -268,9 +268,9 @@
       <url hash="f1cab7c2">2021.naacl-main.18</url>
       <doi>10.18653/v1/2021.naacl-main.18</doi>
       <bibkey>liu-etal-2021-counterfactual</bibkey>
+      <video href="2021.naacl-main.18.mp4"/>
       <pwccode url="https://github.com/marziehf/DataAugmentationNMT" additional="false">marziehf/DataAugmentationNMT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
-      <video href="2021.naacl-main.18.mp4"/>
     </paper>
     <paper id="19">
       <title>Cultural and Geographical Influences on Image Translatability of Words across Languages</title>
@@ -284,9 +284,9 @@
       <url hash="efb60ccb">2021.naacl-main.19</url>
       <doi>10.18653/v1/2021.naacl-main.19</doi>
       <bibkey>khani-etal-2021-cultural</bibkey>
+      <video href="2021.naacl-main.19.mp4"/>
       <pwccode url="https://github.com/nikzadkhani/MMID-CNN-Analysis" additional="false">nikzadkhani/MMID-CNN-Analysis</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mmid">MMID</pwcdataset>
-      <video href="2021.naacl-main.19.mp4"/>
     </paper>
     <paper id="20">
       <title>Multilingual <fixed-case>BERT</fixed-case> Post-Pretraining Alignment</title>
@@ -301,11 +301,11 @@
       <url hash="ab3f0c28">2021.naacl-main.20</url>
       <doi>10.18653/v1/2021.naacl-main.20</doi>
       <bibkey>pan-etal-2021-multilingual</bibkey>
+      <video href="2021.naacl-main.20.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/imagenet">ImageNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.naacl-main.20.mp4"/>
     </paper>
     <paper id="21">
       <title>A Million Tweets Are Worth a Few Points: Tuning Transformers for Customer Service Tasks</title>
@@ -319,8 +319,8 @@
       <url hash="d20f21ee">2021.naacl-main.21</url>
       <doi>10.18653/v1/2021.naacl-main.21</doi>
       <bibkey>hadifar-etal-2021-million</bibkey>
-      <pwccode url="https://github.com/hadifar/customerservicetasks" additional="false">hadifar/customerservicetasks</pwccode>
       <video href="2021.naacl-main.21.mp4"/>
+      <pwccode url="https://github.com/hadifar/customerservicetasks" additional="false">hadifar/customerservicetasks</pwccode>
     </paper>
     <paper id="22">
       <title>Paragraph-level Rationale Extraction through Regularization: A case study on <fixed-case>E</fixed-case>uropean Court of Human Rights Cases</title>
@@ -335,9 +335,9 @@
       <url hash="807ce5fe">2021.naacl-main.22</url>
       <doi>10.18653/v1/2021.naacl-main.22</doi>
       <bibkey>chalkidis-etal-2021-paragraph</bibkey>
+      <video href="2021.naacl-main.22.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ecthr">ECtHR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/echr">ECHR</pwcdataset>
-      <video href="2021.naacl-main.22.mp4"/>
     </paper>
     <paper id="23">
       <title>Answering Product-Questions by Utilizing Questions from Other Contextually Similar Products</title>
@@ -351,8 +351,8 @@
       <url hash="ec1ac4eb">2021.naacl-main.23</url>
       <doi>10.18653/v1/2021.naacl-main.23</doi>
       <bibkey>rozen-etal-2021-answering</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/amazon-pqa">Amazon-PQA</pwcdataset>
       <video href="2021.naacl-main.23.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/amazon-pqa">Amazon-PQA</pwcdataset>
     </paper>
     <paper id="24">
       <title><fixed-case>E</fixed-case>n<fixed-case>S</fixed-case>id<fixed-case>N</fixed-case>et: Enhanced Hybrid <fixed-case>S</fixed-case>iamese-Deep Network for grouping clinical trials into drug-development pathways</title>
@@ -374,9 +374,9 @@
       <url hash="a6a86ea8">2021.naacl-main.25</url>
       <doi>10.18653/v1/2021.naacl-main.25</doi>
       <bibkey>manolache-etal-2021-date</bibkey>
+      <video href="2021.naacl-main.25.mp4"/>
       <pwccode url="https://github.com/bit-ml/date" additional="false">bit-ml/date</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
-      <video href="2021.naacl-main.25.mp4"/>
     </paper>
     <paper id="26">
       <title>A Simple Approach for Handling Out-of-Vocabulary Identifiers in Deep Learning for Source Code</title>
@@ -413,12 +413,12 @@
       <url hash="2bd41961">2021.naacl-main.28</url>
       <doi>10.18653/v1/2021.naacl-main.28</doi>
       <bibkey>thakur-etal-2021-augmented</bibkey>
+      <video href="2021.naacl-main.28.mp4"/>
       <pwccode url="https://github.com/UKPLab/sentence-transformers" additional="false">UKPLab/sentence-transformers</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quora-question-pairs">Quora Question Pairs</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.naacl-main.28.mp4"/>
     </paper>
     <paper id="29">
       <title><fixed-case>S</fixed-case>m<fixed-case>B</fixed-case>o<fixed-case>P</fixed-case>: Semi-autoregressive Bottom-up Semantic Parsing</title>
@@ -429,9 +429,9 @@
       <url hash="922ab628">2021.naacl-main.29</url>
       <doi>10.18653/v1/2021.naacl-main.29</doi>
       <bibkey>rubin-berant-2021-smbop</bibkey>
+      <video href="2021.naacl-main.29.mp4"/>
       <pwccode url="https://github.com/OhadRubin/SmBop" additional="false">OhadRubin/SmBop</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
-      <video href="2021.naacl-main.29.mp4"/>
     </paper>
     <paper id="30">
       <title><fixed-case>SGL</fixed-case>: Speaking the Graph Languages of Semantic Parsing via Multilingual Translation</title>
@@ -456,8 +456,8 @@
       <award>Outstanding Long Paper</award>
       <doi>10.18653/v1/2021.naacl-main.31</doi>
       <bibkey>conia-etal-2021-unifying</bibkey>
-      <pwccode url="https://github.com/SapienzaNLP/unify-srl" additional="false">SapienzaNLP/unify-srl</pwccode>
       <video href="2021.naacl-main.31.mp4"/>
+      <pwccode url="https://github.com/SapienzaNLP/unify-srl" additional="false">SapienzaNLP/unify-srl</pwccode>
     </paper>
     <paper id="32">
       <title>Fool Me Twice: Entailment from <fixed-case>W</fixed-case>ikipedia Gamification</title>
@@ -471,10 +471,10 @@
       <url hash="3f1a7b73">2021.naacl-main.32</url>
       <doi>10.18653/v1/2021.naacl-main.32</doi>
       <bibkey>eisenschlos-etal-2021-fool</bibkey>
+      <video href="2021.naacl-main.32.mp4"/>
       <pwccode url="https://github.com/google-research/fool-me-twice" additional="false">google-research/fool-me-twice</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fm2">FM2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
-      <video href="2021.naacl-main.32.mp4"/>
     </paper>
     <paper id="33">
       <title>Meta-Learning for Domain Generalization in Semantic Parsing</title>
@@ -498,8 +498,8 @@
       <url hash="94289fe4">2021.naacl-main.34</url>
       <doi>10.18653/v1/2021.naacl-main.34</doi>
       <bibkey>schiller-etal-2021-aspect</bibkey>
-      <pwccode url="https://github.com/UKPLab/controlled-argument-generation" additional="false">UKPLab/controlled-argument-generation</pwccode>
       <video href="2021.naacl-main.34.mp4"/>
+      <pwccode url="https://github.com/UKPLab/controlled-argument-generation" additional="false">UKPLab/controlled-argument-generation</pwccode>
     </paper>
     <paper id="35">
       <title>Text Generation from Discourse Representation Structures</title>
@@ -526,8 +526,8 @@
       <url hash="0d0df6e1">2021.naacl-main.36</url>
       <doi>10.18653/v1/2021.naacl-main.36</doi>
       <bibkey>dai-etal-2021-apo</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <video href="2021.naacl-main.36.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
     </paper>
     <paper id="37">
       <title><fixed-case>DART</fixed-case>: Open-Domain Structured Data Record to Text Generation</title>
@@ -560,6 +560,7 @@
       <url hash="5c59fd3f">2021.naacl-main.37</url>
       <doi>10.18653/v1/2021.naacl-main.37</doi>
       <bibkey>nan-etal-2021-dart</bibkey>
+      <video href="2021.naacl-main.37.mp4"/>
       <pwccode url="https://github.com/Yale-LILY/dart" additional="true">Yale-LILY/dart</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dart">DART</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/e2e">E2E</pwcdataset>
@@ -569,7 +570,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/wikibio">WikiBio</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitablequestions">WikiTableQuestions</pwcdataset>
-      <video href="2021.naacl-main.37.mp4"/>
     </paper>
     <paper id="38">
       <title>When Being Unseen from m<fixed-case>BERT</fixed-case> is just the Beginning: Handling New Languages With Multilingual Language Models</title>
@@ -582,9 +582,9 @@
       <url hash="8ecc9794">2021.naacl-main.38</url>
       <doi>10.18653/v1/2021.naacl-main.38</doi>
       <bibkey>muller-etal-2021-unseen</bibkey>
+      <video href="2021.naacl-main.38.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/oscar">OSCAR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikiann-1">WikiAnn</pwcdataset>
-      <video href="2021.naacl-main.38.mp4"/>
     </paper>
     <paper id="39">
       <title>Multi-Adversarial Learning for Cross-Lingual Word Embeddings</title>
@@ -596,8 +596,8 @@
       <url hash="add22690">2021.naacl-main.39</url>
       <doi>10.18653/v1/2021.naacl-main.39</doi>
       <bibkey>wang-etal-2021-multi</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/mldoc">MLDoc</pwcdataset>
       <video href="2021.naacl-main.39.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/mldoc">MLDoc</pwcdataset>
     </paper>
     <paper id="40">
       <title>Multi-view Subword Regularization</title>
@@ -610,11 +610,11 @@
       <attachment type="OptionalSupplementaryData" hash="348c382e">2021.naacl-main.40.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.40</doi>
       <bibkey>wang-etal-2021-multi-view</bibkey>
+      <video href="2021.naacl-main.40.mp4"/>
       <pwccode url="https://github.com/cindyxinyiwang/multiview-subword-regularization" additional="false">cindyxinyiwang/multiview-subword-regularization</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.naacl-main.40.mp4"/>
     </paper>
     <paper id="41">
       <title>m<fixed-case>T</fixed-case>5: A Massively Multilingual Pre-trained Text-to-Text Transformer</title>
@@ -631,6 +631,7 @@
       <url hash="23c9315f">2021.naacl-main.41</url>
       <doi>10.18653/v1/2021.naacl-main.41</doi>
       <bibkey>xue-etal-2021-mt5</bibkey>
+      <video href="2021.naacl-main.41.mp4"/>
       <pwccode url="https://github.com/google-research/multilingual-t5" additional="true">google-research/multilingual-t5</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mc4">mC4</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
@@ -647,7 +648,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/terra">TERRa</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xtreme">XTREME</pwcdataset>
-      <video href="2021.naacl-main.41.mp4"/>
     </paper>
     <paper id="42">
       <title><fixed-case>M</fixed-case>eta<fixed-case>XL</fixed-case>: Meta Representation Transformation for Low-resource Cross-lingual Learning</title>
@@ -675,9 +675,9 @@
       <url hash="1b7f2461">2021.naacl-main.43</url>
       <doi>10.18653/v1/2021.naacl-main.43</doi>
       <bibkey>herzig-etal-2021-open</bibkey>
+      <video href="2021.naacl-main.43.mp4"/>
       <pwccode url="https://github.com/google-research/tapas" additional="false">google-research/tapas</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.naacl-main.43.mp4"/>
     </paper>
     <paper id="44">
       <title>Open-Domain Question Answering Goes Conversational via Question Rewriting</title>
@@ -692,13 +692,13 @@
       <url hash="63f40dbe">2021.naacl-main.44</url>
       <doi>10.18653/v1/2021.naacl-main.44</doi>
       <bibkey>anantha-etal-2021-open</bibkey>
+      <video href="2021.naacl-main.44.mp4"/>
       <pwccode url="https://github.com/apple/ml-qrecc" additional="false">apple/ml-qrecc</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/qrecc">QReCC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/canard">CANARD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/quac">QuAC</pwcdataset>
-      <video href="2021.naacl-main.44.mp4"/>
     </paper>
     <paper id="45">
       <title><fixed-case>QA</fixed-case>-<fixed-case>GNN</fixed-case>: Reasoning with Language Models and Knowledge Graphs for Question Answering</title>
@@ -712,12 +712,12 @@
       <url hash="fee6b099">2021.naacl-main.45</url>
       <doi>10.18653/v1/2021.naacl-main.45</doi>
       <bibkey>yasunaga-etal-2021-qa</bibkey>
+      <video href="2021.naacl-main.45.mp4"/>
       <pwccode url="https://github.com/michiyasunaga/qagnn" additional="true">michiyasunaga/qagnn</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/commonsenseqa">CommonsenseQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/medqa-usmle">MedQA-USMLE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/openbookqa">OpenBookQA</pwcdataset>
-      <video href="2021.naacl-main.45.mp4"/>
     </paper>
     <paper id="46">
       <title><fixed-case>XOR</fixed-case> <fixed-case>QA</fixed-case>: Cross-lingual Open-Retrieval Question Answering</title>
@@ -732,6 +732,7 @@
       <url hash="010faf33">2021.naacl-main.46</url>
       <doi>10.18653/v1/2021.naacl-main.46</doi>
       <bibkey>asai-etal-2021-xor</bibkey>
+      <video href="2021.naacl-main.46.mp4"/>
       <pwccode url="https://github.com/AkariAsai/XORQA" additional="true">AkariAsai/XORQA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/xor-tydi-qa">XOR-TYDI QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mkqa">MKQA</pwcdataset>
@@ -739,7 +740,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.naacl-main.46.mp4"/>
     </paper>
     <paper id="47">
       <title><fixed-case>SPARTA</fixed-case>: Efficient Open-Domain Question Answering via Sparse Transformer Matching Retrieval</title>
@@ -751,6 +751,7 @@
       <url hash="f7c7809e">2021.naacl-main.47</url>
       <doi>10.18653/v1/2021.naacl-main.47</doi>
       <bibkey>zhao-etal-2021-sparta</bibkey>
+      <video href="2021.naacl-main.47.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cmrc">CMRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/drcd">DRCD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
@@ -758,7 +759,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.naacl-main.47.mp4"/>
     </paper>
     <paper id="48">
       <title>Implicitly Abusive Language – What does it actually look like and why are we not getting there?</title>
@@ -818,12 +818,12 @@
       <url hash="1d86e42f">2021.naacl-main.52</url>
       <doi>10.18653/v1/2021.naacl-main.52</doi>
       <bibkey>schuster-etal-2021-get</bibkey>
+      <video href="2021.naacl-main.52.mp4"/>
       <pwccode url="https://github.com/TalSchuster/VitaminC" additional="false">TalSchuster/VitaminC</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/vitaminc">VitaminC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws">PAWS</pwcdataset>
-      <video href="2021.naacl-main.52.mp4"/>
     </paper>
     <paper id="53">
       <title>Representing Numbers in <fixed-case>NLP</fixed-case>: a Survey and a Vision</title>
@@ -836,11 +836,11 @@
       <url hash="7b4ef25d">2021.naacl-main.53</url>
       <doi>10.18653/v1/2021.naacl-main.53</doi>
       <bibkey>thawani-etal-2021-representing</bibkey>
+      <video href="2021.naacl-main.53.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/math">MATH</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/numersense">NumerSense</pwcdataset>
-      <video href="2021.naacl-main.53.mp4"/>
     </paper>
     <paper id="54">
       <title>Extending Multi-Document Summarization Evaluation to the Interactive Setting</title>
@@ -855,8 +855,8 @@
       <url hash="12c9acfa">2021.naacl-main.54</url>
       <doi>10.18653/v1/2021.naacl-main.54</doi>
       <bibkey>shapira-etal-2021-extending</bibkey>
-      <pwccode url="https://github.com/OriShapira/InterExp" additional="false">OriShapira/InterExp</pwccode>
       <video href="2021.naacl-main.54.mp4"/>
+      <pwccode url="https://github.com/OriShapira/InterExp" additional="false">OriShapira/InterExp</pwccode>
     </paper>
     <paper id="55">
       <title>Identifying Helpful Sentences in Product Reviews</title>
@@ -882,11 +882,11 @@
       <url hash="db49b1dc">2021.naacl-main.56</url>
       <doi>10.18653/v1/2021.naacl-main.56</doi>
       <bibkey>liu-etal-2021-noisy</bibkey>
+      <video href="2021.naacl-main.56.mp4"/>
       <pwccode url="https://github.com/nlpyang/NoisySumm" additional="false">nlpyang/NoisySumm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikicatsum">WikiCatSum</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisum">WikiSum</pwcdataset>
-      <video href="2021.naacl-main.56.mp4"/>
     </paper>
     <paper id="57">
       <title>Improving Zero and Few-Shot Abstractive Summarization with Intermediate Fine-tuning and Data Augmentation</title>
@@ -903,10 +903,10 @@
       <url hash="ed256c05">2021.naacl-main.57</url>
       <doi>10.18653/v1/2021.naacl-main.57</doi>
       <bibkey>fabbri-etal-2021-improving</bibkey>
+      <video href="2021.naacl-main.57.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bigpatent">BigPatent</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/new-york-times-annotated-corpus">New York Times Annotated Corpus</pwcdataset>
-      <video href="2021.naacl-main.57.mp4"/>
     </paper>
     <paper id="58">
       <title>Enhancing Factual Consistency of Abstractive Summarization</title>
@@ -962,8 +962,8 @@
       <url hash="84b18320">2021.naacl-main.61</url>
       <doi>10.18653/v1/2021.naacl-main.61</doi>
       <bibkey>paranjape-manning-2021-human</bibkey>
-      <pwccode url="https://github.com/AshwinParanjape/human-like-informative-conversations" additional="false">AshwinParanjape/human-like-informative-conversations</pwccode>
       <video href="2021.naacl-main.61.mp4"/>
+      <pwccode url="https://github.com/AshwinParanjape/human-like-informative-conversations" additional="false">AshwinParanjape/human-like-informative-conversations</pwccode>
     </paper>
     <paper id="62">
       <title>A Comparative Study on Schema-Guided Dialogue State Tracking</title>
@@ -986,9 +986,9 @@
       <url hash="780dbd91">2021.naacl-main.63</url>
       <doi>10.18653/v1/2021.naacl-main.63</doi>
       <bibkey>wu-etal-2021-spoken</bibkey>
+      <video href="2021.naacl-main.63.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.naacl-main.63.mp4"/>
     </paper>
     <paper id="64">
       <title>How to Motivate Your Dragon: Teaching Goal-Driven Agents to Speak and Act in Fantasy Worlds</title>
@@ -1006,8 +1006,8 @@
       <revision id="2" href="2021.naacl-main.64v2" hash="eadf0feb" date="2021-05-25">Updated a citation.</revision>
       <doi>10.18653/v1/2021.naacl-main.64</doi>
       <bibkey>ammanabrolu-etal-2021-motivate</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/light-quests">LIGHT-Quests</pwcdataset>
       <video href="2021.naacl-main.64.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/light-quests">LIGHT-Quests</pwcdataset>
     </paper>
     <paper id="65">
       <title>Linking Entities to Unseen Knowledge Bases with Arbitrary Schemas</title>
@@ -1031,8 +1031,8 @@
       <url hash="d07118b2">2021.naacl-main.66</url>
       <doi>10.18653/v1/2021.naacl-main.66</doi>
       <bibkey>karamanolakis-etal-2021-self</bibkey>
-      <pwccode url="https://github.com/microsoft/ASTRA" additional="false">microsoft/ASTRA</pwccode>
       <video href="2021.naacl-main.66.mp4"/>
+      <pwccode url="https://github.com/microsoft/ASTRA" additional="false">microsoft/ASTRA</pwccode>
     </paper>
     <paper id="67">
       <title>Neural Language Modeling for Contextualized Temporal Graph Generation</title>
@@ -1043,8 +1043,8 @@
       <url hash="3fde681b">2021.naacl-main.67</url>
       <doi>10.18653/v1/2021.naacl-main.67</doi>
       <bibkey>madaan-yang-2021-neural</bibkey>
-      <pwccode url="https://github.com/madaan/temporal-graph-gen" additional="false">madaan/temporal-graph-gen</pwccode>
       <video href="2021.naacl-main.67.mp4"/>
+      <pwccode url="https://github.com/madaan/temporal-graph-gen" additional="false">madaan/temporal-graph-gen</pwccode>
     </paper>
     <paper id="68">
       <title>Probabilistic Box Embeddings for Uncertain Knowledge Graph Reasoning</title>
@@ -1059,9 +1059,9 @@
       <url hash="3064c1a7">2021.naacl-main.68</url>
       <doi>10.18653/v1/2021.naacl-main.68</doi>
       <bibkey>chen-etal-2021-probabilistic</bibkey>
+      <video href="2021.naacl-main.68.mp4"/>
       <pwccode url="https://github.com/stasl0217/beurre" additional="false">stasl0217/beurre</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
-      <video href="2021.naacl-main.68.mp4"/>
     </paper>
     <paper id="69">
       <title>Document-Level Event Argument Extraction by Conditional Generation</title>
@@ -1074,9 +1074,9 @@
       <attachment type="OptionalSupplementaryData" hash="44b6e567">2021.naacl-main.69.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.69</doi>
       <bibkey>li-etal-2021-document</bibkey>
+      <video href="2021.naacl-main.69.mp4"/>
       <pwccode url="https://github.com/raspberryice/gen-arg" additional="false">raspberryice/gen-arg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikievents">WikiEvents</pwcdataset>
-      <video href="2021.naacl-main.69.mp4"/>
     </paper>
     <paper id="70">
       <title>Template Filling with Generative Transformers</title>
@@ -1088,8 +1088,8 @@
       <url hash="ced24048">2021.naacl-main.70</url>
       <doi>10.18653/v1/2021.naacl-main.70</doi>
       <bibkey>du-etal-2021-template</bibkey>
-      <pwccode url="https://github.com/xinyadu/gtt" additional="false">xinyadu/gtt</pwccode>
       <video href="2021.naacl-main.70.mp4"/>
+      <pwccode url="https://github.com/xinyadu/gtt" additional="false">xinyadu/gtt</pwccode>
     </paper>
     <paper id="71">
       <title>Towards Interpreting and Mitigating Shortcut Learning Behavior of <fixed-case>NLU</fixed-case> models</title>
@@ -1106,8 +1106,8 @@
       <url hash="01b6750c">2021.naacl-main.71</url>
       <doi>10.18653/v1/2021.naacl-main.71</doi>
       <bibkey>du-etal-2021-towards</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <video href="2021.naacl-main.71.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
     </paper>
     <paper id="72">
       <title>On Attention Redundancy: A Comprehensive Study</title>
@@ -1135,9 +1135,9 @@
       <url hash="03fc0bc3">2021.naacl-main.73</url>
       <doi>10.18653/v1/2021.naacl-main.73</doi>
       <bibkey>lehman-etal-2021-bert</bibkey>
+      <video href="2021.naacl-main.73.mp4"/>
       <pwccode url="https://github.com/elehman16/exposing_patient_data_release" additional="false">elehman16/exposing_patient_data_release</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mimic-iii">MIMIC-III</pwcdataset>
-      <video href="2021.naacl-main.73.mp4"/>
     </paper>
     <paper id="74">
       <title>Low-Complexity Probing via Finding Subnetworks</title>
@@ -1149,9 +1149,9 @@
       <url hash="b5d87445">2021.naacl-main.74</url>
       <doi>10.18653/v1/2021.naacl-main.74</doi>
       <bibkey>cao-etal-2021-low</bibkey>
+      <video href="2021.naacl-main.74.mp4"/>
       <pwccode url="https://github.com/stevenxcao/subnetwork-probing" additional="false">stevenxcao/subnetwork-probing</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
-      <video href="2021.naacl-main.74.mp4"/>
     </paper>
     <paper id="75">
       <title>An Empirical Comparison of Instance Attribution Methods for <fixed-case>NLP</fixed-case></title>
@@ -1164,9 +1164,9 @@
       <url hash="e872d85f">2021.naacl-main.75</url>
       <doi>10.18653/v1/2021.naacl-main.75</doi>
       <bibkey>pezeshkpour-etal-2021-empirical</bibkey>
+      <video href="2021.naacl-main.75.mp4"/>
       <pwccode url="https://github.com/successar/instance_attributions_NLP" additional="false">successar/instance_attributions_NLP</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.75.mp4"/>
     </paper>
     <paper id="76">
       <title>Generalization in Instruction Following Systems</title>
@@ -1193,9 +1193,9 @@
       <url hash="12e5b521">2021.naacl-main.77</url>
       <doi>10.18653/v1/2021.naacl-main.77</doi>
       <bibkey>sun-etal-2021-lightningdot</bibkey>
+      <video href="2021.naacl-main.77.mp4"/>
       <pwccode url="https://github.com/intersun/LightningDOT" additional="false">intersun/LightningDOT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
-      <video href="2021.naacl-main.77.mp4"/>
     </paper>
     <paper id="78">
       <title>Measuring Social Biases in Grounded Vision and Language Embeddings</title>
@@ -1207,9 +1207,9 @@
       <url hash="1feadaa0">2021.naacl-main.78</url>
       <doi>10.18653/v1/2021.naacl-main.78</doi>
       <bibkey>ross-etal-2021-measuring</bibkey>
+      <video href="2021.naacl-main.78.mp4"/>
       <pwccode url="https://github.com/candacelax/bias-in-vision-and-language" additional="false">candacelax/bias-in-vision-and-language</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
-      <video href="2021.naacl-main.78.mp4"/>
     </paper>
     <paper id="79">
       <title><fixed-case>MTAG</fixed-case>: Modal-Temporal Attention Graph for Unaligned Human Multimodal Language Sequences</title>
@@ -1226,9 +1226,9 @@
       <url hash="f1785926">2021.naacl-main.79</url>
       <doi>10.18653/v1/2021.naacl-main.79</doi>
       <bibkey>yang-etal-2021-mtag</bibkey>
+      <video href="2021.naacl-main.79.mp4"/>
       <pwccode url="https://github.com/jedyang97/MTAG" additional="false">jedyang97/MTAG</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
-      <video href="2021.naacl-main.79.mp4"/>
     </paper>
     <paper id="80">
       <title>Grounding Open-Domain Instructions to Automate Web Support Tasks</title>
@@ -1244,9 +1244,9 @@
       <url hash="29969513">2021.naacl-main.80</url>
       <doi>10.18653/v1/2021.naacl-main.80</doi>
       <bibkey>xu-etal-2021-grounding</bibkey>
+      <video href="2021.naacl-main.80.mp4"/>
       <pwccode url="https://github.com/xnancy/russ" additional="false">xnancy/russ</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/russ-dataset">RUSS Dataset</pwcdataset>
-      <video href="2021.naacl-main.80.mp4"/>
     </paper>
     <paper id="81">
       <title>Modular Networks for Compositional Instruction Following</title>
@@ -1260,8 +1260,8 @@
       <url hash="7c40e780">2021.naacl-main.81</url>
       <doi>10.18653/v1/2021.naacl-main.81</doi>
       <bibkey>corona-etal-2021-modular</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/alfred">ALFRED</pwcdataset>
       <video href="2021.naacl-main.81.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/alfred">ALFRED</pwcdataset>
     </paper>
     <paper id="82">
       <title>Improving Cross-Modal Alignment in Vision Language Navigation via Syntactic Information</title>
@@ -1273,9 +1273,9 @@
       <url hash="c7a6b508">2021.naacl-main.82</url>
       <doi>10.18653/v1/2021.naacl-main.82</doi>
       <bibkey>li-etal-2021-improving</bibkey>
+      <video href="2021.naacl-main.82.mp4"/>
       <pwccode url="https://github.com/jialuli-luka/SyntaxVLN" additional="false">jialuli-luka/SyntaxVLN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/rxr">RxR</pwcdataset>
-      <video href="2021.naacl-main.82.mp4"/>
     </paper>
     <paper id="83">
       <title>Improving Pretrained Models for Zero-shot Multi-label Text Classification through Reinforced Label Hierarchy Reasoning</title>
@@ -1289,9 +1289,9 @@
       <attachment type="OptionalSupplementaryData" hash="6c13013c">2021.naacl-main.83.OptionalSupplementaryData.pdf</attachment>
       <doi>10.18653/v1/2021.naacl-main.83</doi>
       <bibkey>liu-etal-2021-improving</bibkey>
+      <video href="2021.naacl-main.83.mp4"/>
       <pwccode url="https://github.com/layneins/Zero-shot-RLHR" additional="false">layneins/Zero-shot-RLHR</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/web-of-science-dataset">WOS</pwcdataset>
-      <video href="2021.naacl-main.83.mp4"/>
     </paper>
     <paper id="84">
       <title>Fine-Tuning Pre-trained Language Model with Weak Supervision: A Contrastive-Regularized Self-Training Approach</title>
@@ -1306,12 +1306,12 @@
       <url hash="e055f2f3">2021.naacl-main.84</url>
       <doi>10.18653/v1/2021.naacl-main.84</doi>
       <bibkey>yu-etal-2021-fine</bibkey>
+      <video href="2021.naacl-main.84.mp4"/>
       <pwccode url="https://github.com/yueyu1030/COSINE" additional="false">yueyu1030/COSINE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
-      <video href="2021.naacl-main.84.mp4"/>
     </paper>
     <paper id="85">
       <title>Posterior Differential Regularization with f-divergence for Improving Model Robustness</title>
@@ -1325,12 +1325,12 @@
       <url hash="b320aa65">2021.naacl-main.85</url>
       <doi>10.18653/v1/2021.naacl-main.85</doi>
       <bibkey>cheng-etal-2021-posterior</bibkey>
+      <video href="2021.naacl-main.85.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bioasq">BioASQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrqa-2019">MRQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.85.mp4"/>
     </paper>
     <paper id="86">
       <title>Understanding Hard Negatives in Noise Contrastive Estimation</title>
@@ -1341,11 +1341,11 @@
       <url hash="3b6c4213">2021.naacl-main.86</url>
       <doi>10.18653/v1/2021.naacl-main.86</doi>
       <bibkey>zhang-stratos-2021-understanding</bibkey>
+      <video href="2021.naacl-main.86.mp4"/>
       <pwccode url="https://github.com/WenzhengZhang/hard-nce-el" additional="false">WenzhengZhang/hard-nce-el</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/aida-conll-yago">AIDA CoNLL-YAGO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/zeshel">ZESHEL</pwcdataset>
-      <video href="2021.naacl-main.86.mp4"/>
     </paper>
     <paper id="87">
       <title>Certified Robustness to Word Substitution Attack with Differential Privacy</title>
@@ -1359,9 +1359,9 @@
       <attachment type="OptionalSupplementaryData" hash="1ff19c5d">2021.naacl-main.87.OptionalSupplementaryData.pdf</attachment>
       <doi>10.18653/v1/2021.naacl-main.87</doi>
       <bibkey>wang-etal-2021-certified</bibkey>
+      <video href="2021.naacl-main.87.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.naacl-main.87.mp4"/>
     </paper>
     <paper id="88">
       <title><fixed-case>DR</fixed-case>e<fixed-case>C</fixed-case>a: A General Task Augmentation Strategy for Few-Shot Natural Language Inference</title>
@@ -1373,9 +1373,9 @@
       <url hash="ff6f450a">2021.naacl-main.88</url>
       <doi>10.18653/v1/2021.naacl-main.88</doi>
       <bibkey>murty-etal-2021-dreca</bibkey>
+      <video href="2021.naacl-main.88.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
-      <video href="2021.naacl-main.88.mp4"/>
     </paper>
     <paper id="89">
       <title>Harnessing Multilinguality in Unsupervised Machine Translation for Rare Languages</title>
@@ -1388,8 +1388,8 @@
       <url hash="6817eaa9">2021.naacl-main.89</url>
       <doi>10.18653/v1/2021.naacl-main.89</doi>
       <bibkey>garcia-etal-2021-harnessing</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/flores">FLoRes</pwcdataset>
       <video href="2021.naacl-main.89.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/flores">FLoRes</pwcdataset>
     </paper>
     <paper id="90">
       <title>Macro-Average: Rare Types Are Important Too</title>
@@ -1402,9 +1402,9 @@
       <url hash="51571b68">2021.naacl-main.90</url>
       <doi>10.18653/v1/2021.naacl-main.90</doi>
       <bibkey>gowda-etal-2021-macro</bibkey>
+      <video href="2021.naacl-main.90.mp4"/>
       <pwccode url="https://github.com/thammegowda/007-mt-eval-macro" additional="false">thammegowda/007-mt-eval-macro</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/webnlg">WebNLG</pwcdataset>
-      <video href="2021.naacl-main.90.mp4"/>
     </paper>
     <paper id="91">
       <title>Assessing Reference-Free Peer Evaluation for Machine Translation</title>
@@ -1417,8 +1417,8 @@
       <url hash="74221b78">2021.naacl-main.91</url>
       <doi>10.18653/v1/2021.naacl-main.91</doi>
       <bibkey>agrawal-etal-2021-assessing</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wmt19-metrics-task">WMT19 Metrics Task</pwcdataset>
       <video href="2021.naacl-main.91.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wmt19-metrics-task">WMT19 Metrics Task</pwcdataset>
     </paper>
     <paper id="92">
       <title>The Curious Case of Hallucinations in Neural Machine Translation</title>
@@ -1430,8 +1430,8 @@
       <url hash="dc8aead2">2021.naacl-main.92</url>
       <doi>10.18653/v1/2021.naacl-main.92</doi>
       <bibkey>raunak-etal-2021-curious</bibkey>
-      <pwccode url="https://github.com/vyraun/hallucinations" additional="false">vyraun/hallucinations</pwccode>
       <video href="2021.naacl-main.92.mp4"/>
+      <pwccode url="https://github.com/vyraun/hallucinations" additional="false">vyraun/hallucinations</pwccode>
     </paper>
     <paper id="93">
       <title>Towards Continual Learning for Multilingual Machine Translation via Vocabulary Substitution</title>
@@ -1468,11 +1468,11 @@
       <url hash="80057956">2021.naacl-main.95</url>
       <doi>10.18653/v1/2021.naacl-main.95</doi>
       <bibkey>banerjee-etal-2021-self</bibkey>
+      <video href="2021.naacl-main.95.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsqa">NewsQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qa-srl">QA-SRL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.naacl-main.95.mp4"/>
     </paper>
     <paper id="96">
       <title>Capturing Row and Column Semantics in Transformer Based Question Answering over Tables</title>
@@ -1491,12 +1491,12 @@
       <url hash="d46f8805">2021.naacl-main.96</url>
       <doi>10.18653/v1/2021.naacl-main.96</doi>
       <bibkey>glass-etal-2021-capturing</bibkey>
+      <video href="2021.naacl-main.96.mp4"/>
       <pwccode url="https://github.com/IBM/row-column-intersection" additional="false">IBM/row-column-intersection</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitablequestions">WikiTableQuestions</pwcdataset>
-      <video href="2021.naacl-main.96.mp4"/>
     </paper>
     <paper id="97">
       <title>Explainable Multi-hop Verbal Reasoning Through Internal Monologue</title>
@@ -1519,11 +1519,11 @@
       <url hash="4a3ec876">2021.naacl-main.98</url>
       <doi>10.18653/v1/2021.naacl-main.98</doi>
       <bibkey>chen-durrett-2021-robust</bibkey>
+      <video href="2021.naacl-main.98.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bio">Bio</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrqa-2019">MRQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsqa">NewsQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.naacl-main.98.mp4"/>
     </paper>
     <paper id="99">
       <title>Text Modular Networks: Learning to Decompose Tasks in the Language of Existing Models</title>
@@ -1537,12 +1537,12 @@
       <url hash="741d30b5">2021.naacl-main.99</url>
       <doi>10.18653/v1/2021.naacl-main.99</doi>
       <bibkey>khot-etal-2021-text</bibkey>
+      <video href="2021.naacl-main.99.mp4"/>
       <pwccode url="https://github.com/allenai/modularqa" additional="false">allenai/modularqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/break">BREAK</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.naacl-main.99.mp4"/>
     </paper>
     <paper id="100">
       <title><fixed-case>RECONSIDER</fixed-case>: Improved Re-Ranking using Span-Focused Cross-Attention for Open Domain Question Answering</title>
@@ -1555,9 +1555,9 @@
       <url hash="6e6d549d">2021.naacl-main.100</url>
       <doi>10.18653/v1/2021.naacl-main.100</doi>
       <bibkey>iyer-etal-2021-reconsider</bibkey>
+      <video href="2021.naacl-main.100.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.naacl-main.100.mp4"/>
     </paper>
     <paper id="101">
       <title>On the Transferability of Minimal Prediction Preserving Inputs in Question Answering</title>
@@ -1569,12 +1569,12 @@
       <url hash="8715b22d">2021.naacl-main.101</url>
       <doi>10.18653/v1/2021.naacl-main.101</doi>
       <bibkey>longpre-etal-2021-transferability</bibkey>
+      <video href="2021.naacl-main.101.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/newsqa">NewsQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/searchqa">SearchQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.naacl-main.101.mp4"/>
     </paper>
     <paper id="102">
       <title>Understanding by Understanding Not: Modeling Negation in Language Models</title>
@@ -1589,11 +1589,11 @@
       <url hash="b16c5482">2021.naacl-main.102</url>
       <doi>10.18653/v1/2021.naacl-main.102</doi>
       <bibkey>hosseini-etal-2021-understanding</bibkey>
+      <video href="2021.naacl-main.102.mp4"/>
       <pwccode url="https://github.com/arianhosseini/negation-learning" additional="false">arianhosseini/negation-learning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
-      <video href="2021.naacl-main.102.mp4"/>
     </paper>
     <paper id="103">
       <title><fixed-case>D</fixed-case>uo<fixed-case>RAT</fixed-case>: Towards Simpler Text-to-<fixed-case>SQL</fixed-case> Models</title>
@@ -1607,8 +1607,8 @@
       <url hash="65e26d11">2021.naacl-main.103</url>
       <doi>10.18653/v1/2021.naacl-main.103</doi>
       <bibkey>scholak-etal-2021-duorat</bibkey>
-      <pwccode url="https://github.com/ElementAI/duorat" additional="false">ElementAI/duorat</pwccode>
       <video href="2021.naacl-main.103.mp4"/>
+      <pwccode url="https://github.com/ElementAI/duorat" additional="false">ElementAI/duorat</pwccode>
     </paper>
     <paper id="104">
       <title>Looking Beyond Sentence-Level Natural Language Inference for Question Answering and Text Summarization</title>
@@ -1623,13 +1623,13 @@
       <url hash="492ce9df">2021.naacl-main.104</url>
       <doi>10.18653/v1/2021.naacl-main.104</doi>
       <bibkey>mishra-etal-2021-looking</bibkey>
+      <video href="2021.naacl-main.104.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cosmosqa">CosmosQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dream">DREAM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multirc">MultiRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.naacl-main.104.mp4"/>
     </paper>
     <paper id="105">
       <title>Structure-Grounded Pretraining for Text-to-<fixed-case>SQL</fixed-case></title>
@@ -1644,12 +1644,12 @@
       <url hash="708167be">2021.naacl-main.105</url>
       <doi>10.18653/v1/2021.naacl-main.105</doi>
       <bibkey>deng-etal-2021-structure</bibkey>
+      <video href="2021.naacl-main.105.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-realistic">Spider-Realistic</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/totto">ToTTo</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
-      <video href="2021.naacl-main.105.mp4"/>
     </paper>
     <paper id="106">
       <title>Incremental Few-shot Text Classification with Multi-round New Classes: Formulation, Dataset and System</title>
@@ -1689,9 +1689,9 @@
       <url hash="7e0c9ccd">2021.naacl-main.108</url>
       <doi>10.18653/v1/2021.naacl-main.108</doi>
       <bibkey>huang-etal-2021-disentangling</bibkey>
+      <video href="2021.naacl-main.108.mp4"/>
       <pwccode url="https://github.com/uclanlp/ParaBART" additional="false">uclanlp/ParaBART</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/senteval">SentEval</pwcdataset>
-      <video href="2021.naacl-main.108.mp4"/>
     </paper>
     <paper id="109">
       <title>Structure-Aware Abstractive Conversation Summarization via Discourse and Action Graphs</title>
@@ -1702,9 +1702,9 @@
       <url hash="58cc357e">2021.naacl-main.109</url>
       <doi>10.18653/v1/2021.naacl-main.109</doi>
       <bibkey>chen-yang-2021-structure</bibkey>
+      <video href="2021.naacl-main.109.mp4"/>
       <pwccode url="https://github.com/GT-SALT/Structure-Aware-BART" additional="false">GT-SALT/Structure-Aware-BART</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.naacl-main.109.mp4"/>
     </paper>
     <paper id="110">
       <title>A New Approach to Overgenerating and Scoring Abstractive Summaries</title>
@@ -1717,9 +1717,9 @@
       <url hash="1d44cc71">2021.naacl-main.110</url>
       <doi>10.18653/v1/2021.naacl-main.110</doi>
       <bibkey>song-etal-2021-new</bibkey>
+      <video href="2021.naacl-main.110.mp4"/>
       <pwccode url="https://github.com/ucfnlp/varying-length-summ" additional="false">ucfnlp/varying-length-summ</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/newsroom">NEWSROOM</pwcdataset>
-      <video href="2021.naacl-main.110.mp4"/>
     </paper>
     <paper id="111">
       <title><fixed-case>D</fixed-case>2<fixed-case>S</fixed-case>: Document-to-Slide Generation Via Query-Based Text Summarization</title>
@@ -1733,11 +1733,11 @@
       <url hash="19567cf2">2021.naacl-main.111</url>
       <doi>10.18653/v1/2021.naacl-main.111</doi>
       <bibkey>sun-etal-2021-d2s</bibkey>
+      <video href="2021.naacl-main.111.mp4"/>
       <pwccode url="https://github.com/IBM/document2slides" additional="false">IBM/document2slides</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sciduet">SciDuet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/eli5">ELI5</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.naacl-main.111.mp4"/>
     </paper>
     <paper id="112">
       <title>Efficient Attentions for Long Document Summarization</title>
@@ -1751,9 +1751,9 @@
       <url hash="a9d9cd13">2021.naacl-main.112</url>
       <doi>10.18653/v1/2021.naacl-main.112</doi>
       <bibkey>huang-etal-2021-efficient</bibkey>
+      <video href="2021.naacl-main.112.mp4"/>
       <pwccode url="https://github.com/luyang-huang96/LongDocSum" additional="false">luyang-huang96/LongDocSum</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/govreport">GovReport</pwcdataset>
-      <video href="2021.naacl-main.112.mp4"/>
     </paper>
     <paper id="113">
       <title><fixed-case>R</fixed-case>ef<fixed-case>S</fixed-case>um: Refactoring Neural Summarization</title>
@@ -1765,10 +1765,10 @@
       <url hash="e5195109">2021.naacl-main.113</url>
       <doi>10.18653/v1/2021.naacl-main.113</doi>
       <bibkey>liu-etal-2021-refsum</bibkey>
+      <video href="2021.naacl-main.113.mp4"/>
       <pwccode url="https://github.com/yixinL7/Refactoring-Summarization" additional="false">yixinL7/Refactoring-Summarization</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihow">WikiHow</pwcdataset>
-      <video href="2021.naacl-main.113.mp4"/>
     </paper>
     <paper id="114">
       <title>Annotating and Modeling Fine-grained Factuality in Summarization</title>
@@ -1779,9 +1779,9 @@
       <url hash="95db9a84">2021.naacl-main.114</url>
       <doi>10.18653/v1/2021.naacl-main.114</doi>
       <bibkey>goyal-durrett-2021-annotating</bibkey>
+      <video href="2021.naacl-main.114.mp4"/>
       <pwccode url="https://github.com/tagoyal/factuality-datasets" additional="true">tagoyal/factuality-datasets</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
-      <video href="2021.naacl-main.114.mp4"/>
     </paper>
     <paper id="115">
       <title>Larger-Context Tagging: When and Why Does It Work?</title>
@@ -1807,9 +1807,9 @@
       <url hash="33bae495">2021.naacl-main.116</url>
       <doi>10.18653/v1/2021.naacl-main.116</doi>
       <bibkey>li-etal-2021-neural</bibkey>
+      <video href="2021.naacl-main.116.mp4"/>
       <pwccode url="https://github.com/LeePleased/LeftmostSeg" additional="false">LeePleased/LeftmostSeg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
-      <video href="2021.naacl-main.116.mp4"/>
     </paper>
     <paper id="117">
       <title><fixed-case>PCFG</fixed-case>s Can Do Better: Inducing Probabilistic Context-Free Grammars with Many Symbols</title>
@@ -1821,9 +1821,9 @@
       <url hash="a1d06782">2021.naacl-main.117</url>
       <doi>10.18653/v1/2021.naacl-main.117</doi>
       <bibkey>yang-etal-2021-pcfgs</bibkey>
+      <video href="2021.naacl-main.117.mp4"/>
       <pwccode url="https://github.com/sustcsonglin/TN-PCFG" additional="false">sustcsonglin/TN-PCFG</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.naacl-main.117.mp4"/>
     </paper>
     <paper id="118">
       <title><fixed-case>GEMNET</fixed-case>: Effective Gated Gazetteer Representations for Recognizing Complex Entities in Low-context Input</title>
@@ -1836,9 +1836,9 @@
       <url hash="83cc6ffd">2021.naacl-main.118</url>
       <doi>10.18653/v1/2021.naacl-main.118</doi>
       <bibkey>meng-etal-2021-gemnet</bibkey>
+      <video href="2021.naacl-main.118.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wnut-2017-emerging-and-rare-entity">WNUT 2017</pwcdataset>
-      <video href="2021.naacl-main.118.mp4"/>
     </paper>
     <paper id="119">
       <title>Video-aided Unsupervised Grammar Induction</title>
@@ -1855,11 +1855,11 @@
       <award>Best Long Paper</award>
       <doi>10.18653/v1/2021.naacl-main.119</doi>
       <bibkey>zhang-etal-2021-video</bibkey>
+      <video href="2021.naacl-main.119.mp4"/>
       <pwccode url="https://github.com/Sy-Zhang/MMC-PCFG" additional="false">Sy-Zhang/MMC-PCFG</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/didemo">DiDeMo</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/places">Places</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/youcook2">YouCook2</pwcdataset>
-      <video href="2021.naacl-main.119.mp4"/>
     </paper>
     <paper id="120">
       <title>Generating Negative Samples by Manipulating Golden Responses for Unsupervised Learning of a Response Evaluation Model</title>
@@ -1872,11 +1872,11 @@
       <url hash="7397eb94">2021.naacl-main.120</url>
       <doi>10.18653/v1/2021.naacl-main.120</doi>
       <bibkey>park-etal-2021-generating</bibkey>
+      <video href="2021.naacl-main.120.mp4"/>
       <pwccode url="https://github.com/nlpcl-lab/dialog-eval-hard-negative" additional="false">nlpcl-lab/dialog-eval-hard-negative</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog-1">DailyDialog++</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
-      <video href="2021.naacl-main.120.mp4"/>
     </paper>
     <paper id="121">
       <title>How Robust are Fact Checking Systems on Colloquial Claims?</title>
@@ -1889,9 +1889,9 @@
       <url hash="b2357140">2021.naacl-main.121</url>
       <doi>10.18653/v1/2021.naacl-main.121</doi>
       <bibkey>kim-etal-2021-robust</bibkey>
+      <video href="2021.naacl-main.121.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.naacl-main.121.mp4"/>
     </paper>
     <paper id="122">
       <title>Fine-grained Post-training for Improving Retrieval-based Dialogue Systems</title>
@@ -1907,12 +1907,12 @@
       <attachment type="OptionalSupplementaryData" hash="7644b256">2021.naacl-main.122.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.122</doi>
       <bibkey>han-etal-2021-fine</bibkey>
+      <video href="2021.naacl-main.122.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/douban">Douban</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/e-commerce-1">E-commerce</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rrs">RRS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rrs-ranking-test">RRS Ranking Test</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ubuntu-dialogue-corpus">UDC</pwcdataset>
-      <video href="2021.naacl-main.122.mp4"/>
     </paper>
     <paper id="123">
       <title>Put Chatbot into Its Interlocutor’s Shoes: New Framework to Learn Chatbot Responding with Intention</title>
@@ -1945,8 +1945,8 @@
       <url hash="b543a644">2021.naacl-main.124</url>
       <doi>10.18653/v1/2021.naacl-main.124</doi>
       <bibkey>sun-etal-2021-adding</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/sgd">SGD</pwcdataset>
       <video href="2021.naacl-main.124.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/sgd">SGD</pwcdataset>
     </paper>
     <paper id="125">
       <title>Incorporating Syntax and Semantics in Coreference Resolution with Heterogeneous Graph Attention Network</title>
@@ -1957,8 +1957,8 @@
       <url hash="4c7242a7">2021.naacl-main.125</url>
       <doi>10.18653/v1/2021.naacl-main.125</doi>
       <bibkey>jiang-cohn-2021-incorporating</bibkey>
-      <pwccode url="https://github.com/fantabulous-j/coref-hgat" additional="false">fantabulous-j/coref-hgat</pwccode>
       <video href="2021.naacl-main.125.mp4"/>
+      <pwccode url="https://github.com/fantabulous-j/coref-hgat" additional="false">fantabulous-j/coref-hgat</pwccode>
     </paper>
     <paper id="126">
       <title>Context Tracking Network: Graph-based Context Modeling for Implicit Discourse Relation Recognition</title>
@@ -1986,8 +1986,8 @@
       <url hash="cebfff98">2021.naacl-main.127</url>
       <doi>10.18653/v1/2021.naacl-main.127</doi>
       <bibkey>kobayashi-etal-2021-improving</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/rst-dt">RST-DT</pwcdataset>
       <video href="2021.naacl-main.127.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/rst-dt">RST-DT</pwcdataset>
     </paper>
     <paper id="128">
       <title><fixed-case>RST</fixed-case> Parsing from Scratch</title>
@@ -2013,9 +2013,9 @@
       <url hash="dff13f7d">2021.naacl-main.129</url>
       <doi>10.18653/v1/2021.naacl-main.129</doi>
       <bibkey>ferracane-etal-2021-answer</bibkey>
+      <video href="2021.naacl-main.129.mp4"/>
       <pwccode url="https://github.com/elisaF/subjective_discourse" additional="false">elisaF/subjective_discourse</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/subjective-discourse">Subjective Discourse</pwcdataset>
-      <video href="2021.naacl-main.129.mp4"/>
     </paper>
     <paper id="130">
       <title>Evaluating the Impact of a Hierarchical Discourse Representation on Entity Coreference Resolution Performance</title>
@@ -2064,8 +2064,8 @@
       <url hash="8ba37471">2021.naacl-main.133</url>
       <doi>10.18653/v1/2021.naacl-main.133</doi>
       <bibkey>kiegeland-kreutzer-2021-revisiting</bibkey>
-      <pwccode url="https://github.com/samuki/reinforce-joey" additional="false">samuki/reinforce-joey</pwccode>
       <video href="2021.naacl-main.133.mp4"/>
+      <pwccode url="https://github.com/samuki/reinforce-joey" additional="false">samuki/reinforce-joey</pwccode>
     </paper>
     <paper id="134">
       <title>Learning to Organize a Bag of Words into Sentences with Neural Networks: An Empirical Study</title>
@@ -2080,8 +2080,8 @@
       <url hash="ffc26cf2">2021.naacl-main.134</url>
       <doi>10.18653/v1/2021.naacl-main.134</doi>
       <bibkey>tao-etal-2021-learning</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <video href="2021.naacl-main.134.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
     </paper>
     <paper id="135">
       <title>Mask Attention Networks: Rethinking and Strengthen Transformer</title>
@@ -2099,9 +2099,9 @@
       <url hash="720f933a">2021.naacl-main.135</url>
       <doi>10.18653/v1/2021.naacl-main.135</doi>
       <bibkey>fan-etal-2021-mask</bibkey>
+      <video href="2021.naacl-main.135.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
-      <video href="2021.naacl-main.135.mp4"/>
     </paper>
     <paper id="136">
       <title><fixed-case>ERNIE</fixed-case>-Gram: Pre-Training with Explicitly N-Gram Masked Language Modeling for Natural Language Understanding</title>
@@ -2117,6 +2117,7 @@
       <url hash="e9bf68ce">2021.naacl-main.136</url>
       <doi>10.18653/v1/2021.naacl-main.136</doi>
       <bibkey>xiao-etal-2021-ernie</bibkey>
+      <video href="2021.naacl-main.136.mp4"/>
       <pwccode url="https://github.com/PaddlePaddle/PaddleNLP/blob/develop/paddlenlp/transformers/ernie_gram/modeling.py" additional="true">PaddlePaddle/PaddleNLP</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cmrc">CMRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cmrc-2018">CMRC 2018</pwcdataset>
@@ -2130,7 +2131,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.136.mp4"/>
     </paper>
     <paper id="137">
       <title>Lattice-<fixed-case>BERT</fixed-case>: Leveraging Multi-Granularity Representations in <fixed-case>C</fixed-case>hinese Pre-trained Language Models</title>
@@ -2144,9 +2144,9 @@
       <url hash="25f9f321">2021.naacl-main.137</url>
       <doi>10.18653/v1/2021.naacl-main.137</doi>
       <bibkey>lai-etal-2021-lattice</bibkey>
+      <video href="2021.naacl-main.137.mp4"/>
       <pwccode url="https://github.com/alibaba/AliceMind" additional="true">alibaba/AliceMind</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clue">CLUE</pwcdataset>
-      <video href="2021.naacl-main.137.mp4"/>
     </paper>
     <paper id="138">
       <title>Modeling Event Plausibility with Consistent Conceptual Abstraction</title>
@@ -2159,8 +2159,8 @@
       <url hash="1a571897">2021.naacl-main.138</url>
       <doi>10.18653/v1/2021.naacl-main.138</doi>
       <bibkey>porada-etal-2021-modeling</bibkey>
-      <pwccode url="https://github.com/ianporada/modeling_event_plausibility" additional="false">ianporada/modeling_event_plausibility</pwccode>
       <video href="2021.naacl-main.138.mp4"/>
+      <pwccode url="https://github.com/ianporada/modeling_event_plausibility" additional="false">ianporada/modeling_event_plausibility</pwccode>
     </paper>
     <paper id="139">
       <title><fixed-case>U</fixed-case>mls<fixed-case>BERT</fixed-case>: Clinical Domain Knowledge Augmentation of Contextual Embeddings Using the <fixed-case>U</fixed-case>nified <fixed-case>M</fixed-case>edical <fixed-case>L</fixed-case>anguage <fixed-case>S</fixed-case>ystem <fixed-case>M</fixed-case>etathesaurus</title>
@@ -2174,8 +2174,8 @@
       <url hash="bb95c792">2021.naacl-main.139</url>
       <doi>10.18653/v1/2021.naacl-main.139</doi>
       <bibkey>michalopoulos-etal-2021-umlsbert</bibkey>
-      <pwccode url="https://github.com/gmichalo/UmlsBERT" additional="false">gmichalo/UmlsBERT</pwccode>
       <video href="2021.naacl-main.139.mp4"/>
+      <pwccode url="https://github.com/gmichalo/UmlsBERT" additional="false">gmichalo/UmlsBERT</pwccode>
     </paper>
     <paper id="140">
       <title>Field Embedding: A Unified Grain-Based Framework for Word Representation</title>
@@ -2208,8 +2208,8 @@
       <url hash="d0a22c32">2021.naacl-main.141</url>
       <doi>10.18653/v1/2021.naacl-main.141</doi>
       <bibkey>choi-etal-2021-melbert</bibkey>
-      <pwccode url="https://github.com/jin530/MelBERT" additional="false">jin530/MelBERT</pwccode>
       <video href="2021.naacl-main.141.mp4"/>
+      <pwccode url="https://github.com/jin530/MelBERT" additional="false">jin530/MelBERT</pwccode>
     </paper>
     <paper id="142">
       <title>Non-Parametric Few-Shot Learning for Word Sense Disambiguation</title>
@@ -2221,9 +2221,9 @@
       <url hash="2c8abd0f">2021.naacl-main.142</url>
       <doi>10.18653/v1/2021.naacl-main.142</doi>
       <bibkey>chen-etal-2021-non</bibkey>
+      <video href="2021.naacl-main.142.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/metric-wsd" additional="false">princeton-nlp/metric-wsd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
-      <video href="2021.naacl-main.142.mp4"/>
     </paper>
     <paper id="143">
       <title>Why Do Document-Level Polarity Classifiers Fail?</title>
@@ -2235,8 +2235,8 @@
       <url hash="5f282cc5">2021.naacl-main.143</url>
       <doi>10.18653/v1/2021.naacl-main.143</doi>
       <bibkey>martins-etal-2021-document</bibkey>
-      <pwccode url="https://github.com/karenstemartins/NAACL2021" additional="false">karenstemartins/NAACL2021</pwccode>
       <video href="2021.naacl-main.143.mp4"/>
+      <pwccode url="https://github.com/karenstemartins/NAACL2021" additional="false">karenstemartins/NAACL2021</pwccode>
     </paper>
     <paper id="144">
       <title>A Unified Span-Based Approach for Opinion Mining with Syntactic Constituents</title>
@@ -2253,10 +2253,10 @@
       <url hash="f3512cff">2021.naacl-main.144</url>
       <doi>10.18653/v1/2021.naacl-main.144</doi>
       <bibkey>xia-etal-2021-unified</bibkey>
+      <video href="2021.naacl-main.144.mp4"/>
       <pwccode url="https://github.com/KiroSummer/opinion_mining_with_syn_cons" additional="false">KiroSummer/opinion_mining_with_syn_cons</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.naacl-main.144.mp4"/>
     </paper>
     <paper id="145">
       <title>Target-specified Sequence Labeling with Multi-head Self-attention for Target-oriented Opinion Words Extraction</title>
@@ -2270,8 +2270,8 @@
       <url hash="9759adc5">2021.naacl-main.145</url>
       <doi>10.18653/v1/2021.naacl-main.145</doi>
       <bibkey>feng-etal-2021-target</bibkey>
-      <pwccode url="https://github.com/fengyh3/TSMSA" additional="false">fengyh3/TSMSA</pwccode>
       <video href="2021.naacl-main.145.mp4"/>
+      <pwccode url="https://github.com/fengyh3/TSMSA" additional="false">fengyh3/TSMSA</pwccode>
     </paper>
     <paper id="146">
       <title>Does syntax matter? A strong baseline for Aspect-based Sentiment Analysis with <fixed-case>R</fixed-case>o<fixed-case>BERT</fixed-case>a</title>
@@ -2285,9 +2285,9 @@
       <url hash="56346654">2021.naacl-main.146</url>
       <doi>10.18653/v1/2021.naacl-main.146</doi>
       <bibkey>dai-etal-2021-syntax</bibkey>
+      <video href="2021.naacl-main.146.mp4"/>
       <pwccode url="https://github.com/ROGERDJQ/RoBERTaABSA" additional="false">ROGERDJQ/RoBERTaABSA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/semeval-2014-task-4-sub-task-2">SemEval 2014 Task 4 Sub Task 2</pwcdataset>
-      <video href="2021.naacl-main.146.mp4"/>
     </paper>
     <paper id="147">
       <title>Domain Divergences: A Survey and Empirical Analysis</title>
@@ -2336,8 +2336,8 @@
       <url hash="eec3a5ba">2021.naacl-main.150</url>
       <doi>10.18653/v1/2021.naacl-main.150</doi>
       <bibkey>inaguma-etal-2021-source</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
       <video href="2021.naacl-main.150.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/must-c">MuST-C</pwcdataset>
     </paper>
     <paper id="151">
       <title>Searchable Hidden Intermediates for End-to-End Models of Decomposable Sequence Tasks</title>
@@ -2363,12 +2363,12 @@
       <url hash="3b558c6b">2021.naacl-main.152</url>
       <doi>10.18653/v1/2021.naacl-main.152</doi>
       <bibkey>chung-etal-2021-splat</bibkey>
+      <video href="2021.naacl-main.152.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cmu-mosei">CMU-MOSEI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fluent-speech-commands">Fluent Speech Commands</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/spoken-squad">Spoken-SQuAD</pwcdataset>
-      <video href="2021.naacl-main.152.mp4"/>
     </paper>
     <paper id="153">
       <title>Worldly Wise (<fixed-case>W</fixed-case>o<fixed-case>W</fixed-case>) - Cross-Lingual Knowledge Fusion for Fact-based Visual Spoken-Question Answering</title>
@@ -2383,11 +2383,11 @@
       <attachment type="OptionalSupplementaryData" hash="f66c37cb">2021.naacl-main.153.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.153</doi>
       <bibkey>ramnath-etal-2021-worldly</bibkey>
+      <video href="2021.naacl-main.153.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dbpedia">DBpedia</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/places">Places</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/simplequestions">SimpleQuestions</pwcdataset>
-      <video href="2021.naacl-main.153.mp4"/>
     </paper>
     <paper id="154">
       <title>Align-Refine: Non-Autoregressive Speech Recognition via Iterative Realignment</title>
@@ -2399,8 +2399,8 @@
       <url hash="5b63d4e1">2021.naacl-main.154</url>
       <doi>10.18653/v1/2021.naacl-main.154</doi>
       <bibkey>chi-etal-2021-align</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
       <video href="2021.naacl-main.154.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/librispeech">LibriSpeech</pwcdataset>
     </paper>
     <paper id="155">
       <title>Everything Has a Cause: Leveraging Causal Inference in Legal Text Analysis</title>
@@ -2414,8 +2414,8 @@
       <url hash="37f1a6ce">2021.naacl-main.155</url>
       <doi>10.18653/v1/2021.naacl-main.155</doi>
       <bibkey>liu-etal-2021-everything</bibkey>
-      <pwccode url="https://github.com/xxxiaol/GCI" additional="false">xxxiaol/GCI</pwccode>
       <video href="2021.naacl-main.155.mp4"/>
+      <pwccode url="https://github.com/xxxiaol/GCI" additional="false">xxxiaol/GCI</pwccode>
     </paper>
     <paper id="156">
       <title>Counterfactual Supporting Facts Extraction for Explainable Medical Record Based Diagnosis with Graph Network</title>
@@ -2428,8 +2428,8 @@
       <url hash="aedd8ee0">2021.naacl-main.156</url>
       <doi>10.18653/v1/2021.naacl-main.156</doi>
       <bibkey>wu-etal-2021-counterfactual</bibkey>
-      <pwccode url="https://github.com/ckre/cmge" additional="false">ckre/cmge</pwccode>
       <video href="2021.naacl-main.156.mp4"/>
+      <pwccode url="https://github.com/ckre/cmge" additional="false">ckre/cmge</pwccode>
     </paper>
     <paper id="157">
       <title>Personalized Response Generation via Generative Split Memory Network</title>
@@ -2442,11 +2442,11 @@
       <attachment type="OptionalSupplementaryData" hash="c9ae8735">2021.naacl-main.157.OptionalSupplementaryData.pdf</attachment>
       <doi>10.18653/v1/2021.naacl-main.157</doi>
       <bibkey>wu-etal-2021-personalized</bibkey>
+      <video href="2021.naacl-main.157.mp4"/>
       <pwccode url="https://github.com/willyoung2017/per-chat" additional="false">willyoung2017/per-chat</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/pec">PEC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/persona-chat-1">PERSONA-CHAT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/personaldialog">PersonalDialog</pwcdataset>
-      <video href="2021.naacl-main.157.mp4"/>
     </paper>
     <paper id="158">
       <title>Towards Few-shot Fact-Checking via Perplexity</title>
@@ -2459,8 +2459,8 @@
       <url hash="2a26b242">2021.naacl-main.158</url>
       <doi>10.18653/v1/2021.naacl-main.158</doi>
       <bibkey>lee-etal-2021-towards</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <video href="2021.naacl-main.158.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
     </paper>
     <paper id="159">
       <title>Active<tex-math>^2</tex-math> Learning: Actively reducing redundancies in Active Learning methods for Sequence Tagging and Machine Translation</title>
@@ -2501,9 +2501,9 @@
       <url hash="39018058">2021.naacl-main.161</url>
       <doi>10.18653/v1/2021.naacl-main.161</doi>
       <bibkey>he-etal-2021-model</bibkey>
+      <video href="2021.naacl-main.161.mp4"/>
       <pwccode url="https://github.com/xlhex/extract_and_transfer" additional="false">xlhex/extract_and_transfer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
-      <video href="2021.naacl-main.161.mp4"/>
     </paper>
     <paper id="162">
       <title>A Global Past-Future Early Exit Method for Accelerating Inference of Pre-trained Language Models</title>
@@ -2518,10 +2518,10 @@
       <url hash="e65c760c">2021.naacl-main.162</url>
       <doi>10.18653/v1/2021.naacl-main.162</doi>
       <bibkey>liao-etal-2021-global</bibkey>
+      <video href="2021.naacl-main.162.mp4"/>
       <pwccode url="https://github.com/lancopku/early-exit" additional="false">lancopku/early-exit</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.naacl-main.162.mp4"/>
     </paper>
     <paper id="163">
       <title>Masked Conditional Random Fields for Sequence Labeling</title>
@@ -2534,10 +2534,10 @@
       <url hash="60cfc4ca">2021.naacl-main.163</url>
       <doi>10.18653/v1/2021.naacl-main.163</doi>
       <bibkey>wei-etal-2021-masked</bibkey>
+      <video href="2021.naacl-main.163.mp4"/>
       <pwccode url="https://github.com/DandyQi/MaskedCRF" additional="true">DandyQi/MaskedCRF</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
-      <video href="2021.naacl-main.163.mp4"/>
     </paper>
     <paper id="164">
       <title>Heterogeneous Graph Neural Networks for Concept Prerequisite Relation Learning in Educational Data</title>
@@ -2551,8 +2551,8 @@
       <url hash="1d1d1224">2021.naacl-main.164</url>
       <doi>10.18653/v1/2021.naacl-main.164</doi>
       <bibkey>jia-etal-2021-heterogeneous</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/lecturebank">LectureBank</pwcdataset>
       <video href="2021.naacl-main.164.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/lecturebank">LectureBank</pwcdataset>
     </paper>
     <paper id="165">
       <title>Be Careful about Poisoned Word Embeddings: Exploring the Vulnerability of the Embedding Layers in <fixed-case>NLP</fixed-case> Models</title>
@@ -2567,11 +2567,12 @@
       <url hash="d4506bd6">2021.naacl-main.165</url>
       <doi>10.18653/v1/2021.naacl-main.165</doi>
       <bibkey>yang-etal-2021-careful</bibkey>
+      <video href="2021.naacl-main.165.mp4"/>
       <pwccode url="https://github.com/lancopku/Embedding-Poisoning" additional="false">lancopku/Embedding-Poisoning</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.naacl-main.165.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="166">
       <title><fixed-case>DA</fixed-case>-Transformer: Distance-aware Transformer</title>
@@ -2583,10 +2584,10 @@
       <url hash="a6ea20ce">2021.naacl-main.166</url>
       <doi>10.18653/v1/2021.naacl-main.166</doi>
       <bibkey>wu-etal-2021-da</bibkey>
+      <video href="2021.naacl-main.166.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/mind">MIND</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.166.mp4"/>
     </paper>
     <paper id="167">
       <title><fixed-case>ASAP</fixed-case>: A <fixed-case>C</fixed-case>hinese Review Dataset Towards Aspect Category Sentiment Analysis and Rating Prediction</title>
@@ -2603,8 +2604,8 @@
       <attachment type="OptionalSupplementaryData" hash="ed7b3ad8">2021.naacl-main.167.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.167</doi>
       <bibkey>bu-etal-2021-asap</bibkey>
-      <pwccode url="https://github.com/Meituan-Dianping/asap" additional="false">Meituan-Dianping/asap</pwccode>
       <video href="2021.naacl-main.167.mp4"/>
+      <pwccode url="https://github.com/Meituan-Dianping/asap" additional="false">Meituan-Dianping/asap</pwccode>
     </paper>
     <paper id="168">
       <title>Are <fixed-case>NLP</fixed-case> Models really able to Solve Simple Math Word Problems?</title>
@@ -2618,10 +2619,10 @@
       <attachment type="OptionalSupplementaryData" hash="af0d277e">2021.naacl-main.168.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.168</doi>
       <bibkey>patel-etal-2021-nlp</bibkey>
+      <video href="2021.naacl-main.168.mp4"/>
       <pwccode url="https://github.com/arkilpatel/SVAMP" additional="false">arkilpatel/SVAMP</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/svamp">SVAMP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/math23k">Math23K</pwcdataset>
-      <video href="2021.naacl-main.168.mp4"/>
     </paper>
     <paper id="169">
       <title><fixed-case>WRIME</fixed-case>: A New Dataset for Emotional Intensity Estimation with Subjective and Objective Annotations</title>
@@ -2635,10 +2636,10 @@
       <url hash="8d459ead">2021.naacl-main.169</url>
       <doi>10.18653/v1/2021.naacl-main.169</doi>
       <bibkey>kajiwara-etal-2021-wrime</bibkey>
+      <video href="2021.naacl-main.169.mp4"/>
       <pwccode url="https://github.com/ids-cv/wrime" additional="false">ids-cv/wrime</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/emobank">EmoBank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/isear">ISEAR</pwcdataset>
-      <video href="2021.naacl-main.169.mp4"/>
     </paper>
     <paper id="170">
       <title><fixed-case>KPQA</fixed-case>: A Metric for Generative Question Answering Using Keyphrase Weights</title>
@@ -2655,12 +2656,12 @@
       <attachment type="OptionalSupplementaryData" hash="5a65429a">2021.naacl-main.170.OptionalSupplementaryData.pdf</attachment>
       <doi>10.18653/v1/2021.naacl-main.170</doi>
       <bibkey>lee-etal-2021-kpqa</bibkey>
+      <video href="2021.naacl-main.170.mp4"/>
       <pwccode url="https://github.com/hwanheelee1993/KPQA" additional="false">hwanheelee1993/KPQA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gqa">GQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/narrativeqa">NarrativeQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.naacl-main.170.mp4"/>
     </paper>
     <paper id="171">
       <title><fixed-case>S</fixed-case>tyle<fixed-case>PTB</fixed-case>: A Compositional Benchmark for Fine-grained Controllable Text Style Transfer</title>
@@ -2678,11 +2679,11 @@
       <attachment type="OptionalSupplementaryData" hash="302e28a4">2021.naacl-main.171.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.171</doi>
       <bibkey>lyu-etal-2021-styleptb</bibkey>
+      <video href="2021.naacl-main.171.mp4"/>
       <pwccode url="https://github.com/lvyiwei1/StylePTB" additional="true">lvyiwei1/StylePTB</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/styleptb">StylePTB</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.naacl-main.171.mp4"/>
     </paper>
     <paper id="172">
       <title>Blow the Dog Whistle: A <fixed-case>C</fixed-case>hinese Dataset for Cant Understanding with Common Sense and World Knowledge</title>
@@ -2697,10 +2698,10 @@
       <url hash="074a7e93">2021.naacl-main.172</url>
       <doi>10.18653/v1/2021.naacl-main.172</doi>
       <bibkey>xu-etal-2021-blow</bibkey>
+      <video href="2021.naacl-main.172.mp4"/>
       <pwccode url="https://github.com/JetRunner/dogwhistle" additional="false">JetRunner/dogwhistle</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dogwhistle">DogWhistle</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/clue">CLUE</pwcdataset>
-      <video href="2021.naacl-main.172.mp4"/>
     </paper>
     <paper id="173">
       <title><fixed-case>COVID</fixed-case>-19 Named Entity Recognition for <fixed-case>V</fixed-case>ietnamese</title>
@@ -2712,9 +2713,9 @@
       <url hash="c033111a">2021.naacl-main.173</url>
       <doi>10.18653/v1/2021.naacl-main.173</doi>
       <bibkey>truong-etal-2021-covid</bibkey>
+      <video href="2021.naacl-main.173.mp4"/>
       <pwccode url="https://github.com/VinAIResearch/PhoNER_COVID19" additional="false">VinAIResearch/PhoNER_COVID19</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/phoner-covid19">PhoNER COVID19</pwcdataset>
-      <video href="2021.naacl-main.173.mp4"/>
     </paper>
     <paper id="174">
       <title>Framing Unpacked: A Semi-Supervised Interpretable Multi-View Model of Media Frames</title>
@@ -2728,8 +2729,8 @@
       <url hash="88657dd6">2021.naacl-main.174</url>
       <doi>10.18653/v1/2021.naacl-main.174</doi>
       <bibkey>khanehzar-etal-2021-framing</bibkey>
-      <pwccode url="https://github.com/shinyemimalef/FRISS" additional="false">shinyemimalef/FRISS</pwccode>
       <video href="2021.naacl-main.174.mp4"/>
+      <pwccode url="https://github.com/shinyemimalef/FRISS" additional="false">shinyemimalef/FRISS</pwccode>
     </paper>
     <paper id="175">
       <title>Automatic Classification of Neutralization Techniques in the Narrative of Climate Change Scepticism</title>
@@ -2780,8 +2781,8 @@
       <url hash="ff6bdd78">2021.naacl-main.178</url>
       <doi>10.18653/v1/2021.naacl-main.178</doi>
       <bibkey>zhu-jurgens-2021-structure</bibkey>
-      <pwccode url="https://github.com/lingjzhu/reddit_network" additional="false">lingjzhu/reddit_network</pwccode>
       <video href="2021.naacl-main.178.mp4"/>
+      <pwccode url="https://github.com/lingjzhu/reddit_network" additional="false">lingjzhu/reddit_network</pwccode>
     </paper>
     <paper id="179">
       <title>Modeling Framing in Immigration Discourse on Social Media</title>
@@ -2793,8 +2794,8 @@
       <url hash="3b8307d5">2021.naacl-main.179</url>
       <doi>10.18653/v1/2021.naacl-main.179</doi>
       <bibkey>mendelsohn-etal-2021-modeling</bibkey>
-      <pwccode url="https://github.com/juliamendelsohn/framing" additional="false">juliamendelsohn/framing</pwccode>
       <video href="2021.naacl-main.179.mp4"/>
+      <pwccode url="https://github.com/juliamendelsohn/framing" additional="false">juliamendelsohn/framing</pwccode>
     </paper>
     <paper id="180">
       <title>Modeling the Severity of Complaints in Social Media</title>
@@ -2805,8 +2806,8 @@
       <url hash="bf3bbfeb">2021.naacl-main.180</url>
       <doi>10.18653/v1/2021.naacl-main.180</doi>
       <bibkey>jin-aletras-2021-modeling</bibkey>
-      <pwccode url="https://github.com/mali726/Complaint-Severity" additional="false">mali726/Complaint-Severity</pwccode>
       <video href="2021.naacl-main.180.mp4"/>
+      <pwccode url="https://github.com/mali726/Complaint-Severity" additional="false">mali726/Complaint-Severity</pwccode>
     </paper>
     <paper id="181">
       <title>What About the Precedent: An Information-Theoretic Analysis of Common Law</title>
@@ -2820,9 +2821,9 @@
       <url hash="8795a304">2021.naacl-main.181</url>
       <doi>10.18653/v1/2021.naacl-main.181</doi>
       <bibkey>valvoda-etal-2021-precedent</bibkey>
+      <video href="2021.naacl-main.181.mp4"/>
       <pwccode url="https://github.com/valvoda/Precedent" additional="false">valvoda/Precedent</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/echr">ECHR</pwcdataset>
-      <video href="2021.naacl-main.181.mp4"/>
     </paper>
     <paper id="182">
       <title>Introducing <fixed-case>CAD</fixed-case>: the Contextual Abuse Dataset</title>
@@ -2836,8 +2837,8 @@
       <url hash="92a564b8">2021.naacl-main.182</url>
       <doi>10.18653/v1/2021.naacl-main.182</doi>
       <bibkey>vidgen-etal-2021-introducing</bibkey>
-      <pwccode url="https://github.com/dongpng/cad_naacl2021" additional="false">dongpng/cad_naacl2021</pwccode>
       <video href="2021.naacl-main.182.mp4"/>
+      <pwccode url="https://github.com/dongpng/cad_naacl2021" additional="false">dongpng/cad_naacl2021</pwccode>
     </paper>
     <paper id="183">
       <title>Lifelong Learning of Hate Speech Classification on Social Media</title>
@@ -2877,6 +2878,7 @@
       <award>Outstanding Long Paper</award>
       <doi>10.18653/v1/2021.naacl-main.185</doi>
       <bibkey>schick-schutze-2021-just</bibkey>
+      <video href="2021.naacl-main.185.mp4"/>
       <pwccode url="https://github.com/timoschick/pet" additional="true">timoschick/pet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewglue">FewGlue</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
@@ -2886,7 +2888,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wic">WiC</pwcdataset>
-      <video href="2021.naacl-main.185.mp4"/>
     </paper>
     <paper id="186">
       <title>Static Embeddings as Efficient Knowledge Bases?</title>
@@ -2898,10 +2899,10 @@
       <url hash="ae75df19">2021.naacl-main.186</url>
       <doi>10.18653/v1/2021.naacl-main.186</doi>
       <bibkey>dufter-etal-2021-static</bibkey>
+      <video href="2021.naacl-main.186.mp4"/>
       <pwccode url="https://github.com/pdufter/staticlama" additional="false">pdufter/staticlama</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
-      <video href="2021.naacl-main.186.mp4"/>
     </paper>
     <paper id="187">
       <title>Highly Efficient Knowledge Graph Embedding Learning with <fixed-case>O</fixed-case>rthogonal <fixed-case>P</fixed-case>rocrustes <fixed-case>A</fixed-case>nalysis</title>
@@ -2914,9 +2915,9 @@
       <url hash="66ad8e14">2021.naacl-main.187</url>
       <doi>10.18653/v1/2021.naacl-main.187</doi>
       <bibkey>peng-etal-2021-highly</bibkey>
+      <video href="2021.naacl-main.187.mp4"/>
       <pwccode url="https://github.com/Pzoom522/ProcrustEs-KGE" additional="false">Pzoom522/ProcrustEs-KGE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fb15k-237">FB15k-237</pwcdataset>
-      <video href="2021.naacl-main.187.mp4"/>
     </paper>
     <paper id="188">
       <title>Rethinking Network Pruning – under the Pre-train and Fine-tune Paradigm</title>
@@ -2930,11 +2931,11 @@
       <attachment type="OptionalSupplementaryCode" hash="e435a5d9">2021.naacl-main.188.OptionalSupplementaryCode.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.188</doi>
       <bibkey>xu-etal-2021-rethinking</bibkey>
+      <video href="2021.naacl-main.188.mp4"/>
       <pwccode url="https://github.com/derronxu/sparsebert" additional="false">derronxu/sparsebert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
-      <video href="2021.naacl-main.188.mp4"/>
     </paper>
     <paper id="189">
       <title>Towards a Comprehensive Understanding and Accurate Evaluation of Societal Biases in Pre-Trained Transformers</title>
@@ -2946,8 +2947,8 @@
       <url hash="507cd665">2021.naacl-main.189</url>
       <doi>10.18653/v1/2021.naacl-main.189</doi>
       <bibkey>silva-etal-2021-towards</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/swag">SWAG</pwcdataset>
       <video href="2021.naacl-main.189.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/swag">SWAG</pwcdataset>
     </paper>
     <paper id="190">
       <title>Detoxifying Language Models Risks Marginalizing Minority Voices</title>
@@ -2962,8 +2963,8 @@
       <url hash="b7430224">2021.naacl-main.190</url>
       <doi>10.18653/v1/2021.naacl-main.190</doi>
       <bibkey>xu-etal-2021-detoxifying</bibkey>
-      <pwccode url="https://github.com/albertkx/detoxifying-lms" additional="false">albertkx/detoxifying-lms</pwccode>
       <video href="2021.naacl-main.190.mp4"/>
+      <pwccode url="https://github.com/albertkx/detoxifying-lms" additional="false">albertkx/detoxifying-lms</pwccode>
     </paper>
     <paper id="191">
       <title><fixed-case>HONEST</fixed-case>: Measuring Hurtful Sentence Completion in Language Models</title>
@@ -2975,8 +2976,8 @@
       <url hash="fc53b3e0">2021.naacl-main.191</url>
       <doi>10.18653/v1/2021.naacl-main.191</doi>
       <bibkey>nozza-etal-2021-honest</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/honest-en">HONEST</pwcdataset>
       <video href="2021.naacl-main.191.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/honest-en">HONEST</pwcdataset>
     </paper>
     <paper id="192">
       <title><fixed-case>E</fixed-case>a<fixed-case>S</fixed-case>e: A Diagnostic Tool for <fixed-case>VQA</fixed-case> based on Answer Diversity</title>
@@ -2989,10 +2990,10 @@
       <attachment type="OptionalSupplementaryData" hash="7fe1c15c">2021.naacl-main.192.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.192</doi>
       <bibkey>jolly-etal-2021-ease</bibkey>
+      <video href="2021.naacl-main.192.mp4"/>
       <pwccode url="https://github.com/shailzajolly/ease" additional="false">shailzajolly/ease</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering-v2-0">Visual Question Answering v2.0</pwcdataset>
-      <video href="2021.naacl-main.192.mp4"/>
     </paper>
     <paper id="193">
       <title><fixed-case>D</fixed-case>e<fixed-case>CEMBERT</fixed-case>: Learning from Noisy Instructional Videos via Dense Captions and Entropy Minimization</title>
@@ -3004,10 +3005,10 @@
       <url hash="1473810e">2021.naacl-main.193</url>
       <doi>10.18653/v1/2021.naacl-main.193</doi>
       <bibkey>tang-etal-2021-decembert</bibkey>
+      <video href="2021.naacl-main.193.mp4"/>
       <pwccode url="https://github.com/zinengtang/decembert" additional="false">zinengtang/decembert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/howto100m">HowTo100M</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/youcook2">YouCook2</pwcdataset>
-      <video href="2021.naacl-main.193.mp4"/>
     </paper>
     <paper id="194">
       <title>Improving Generation and Evaluation of Visual Stories via Semantic Consistency</title>
@@ -3019,8 +3020,8 @@
       <url hash="61107827">2021.naacl-main.194</url>
       <doi>10.18653/v1/2021.naacl-main.194</doi>
       <bibkey>maharana-etal-2021-improving</bibkey>
-      <pwccode url="https://github.com/adymaharana/StoryViz" additional="false">adymaharana/StoryViz</pwccode>
       <video href="2021.naacl-main.194.mp4"/>
+      <pwccode url="https://github.com/adymaharana/StoryViz" additional="false">adymaharana/StoryViz</pwccode>
     </paper>
     <paper id="195">
       <title>Multilingual Multimodal Pre-training for Zero-Shot Cross-Lingual Transfer of Vision-Language Models</title>
@@ -3036,10 +3037,10 @@
       <attachment type="OptionalSupplementaryData" hash="0546d73d">2021.naacl-main.195.OptionalSupplementaryData.txt</attachment>
       <doi>10.18653/v1/2021.naacl-main.195</doi>
       <bibkey>huang-etal-2021-multilingual</bibkey>
+      <video href="2021.naacl-main.195.mp4"/>
       <pwccode url="https://github.com/berniebear/Multi-HT100M" additional="false">berniebear/Multi-HT100M</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/howto100m">HowTo100M</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/vatex">VATEX</pwcdataset>
-      <video href="2021.naacl-main.195.mp4"/>
     </paper>
     <paper id="196">
       <title>Video Question Answering with Phrases via Semantic Roles</title>
@@ -3051,9 +3052,9 @@
       <url hash="41bb2b6a">2021.naacl-main.196</url>
       <doi>10.18653/v1/2021.naacl-main.196</doi>
       <bibkey>sadhu-etal-2021-video</bibkey>
+      <video href="2021.naacl-main.196.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/activitynet-qa">ActivityNet-QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/charades">Charades</pwcdataset>
-      <video href="2021.naacl-main.196.mp4"/>
     </paper>
     <paper id="197">
       <title>From Masked Language Modeling to Translation: Non-<fixed-case>E</fixed-case>nglish Auxiliary Tasks Improve Zero-shot Spoken Language Understanding</title>
@@ -3071,12 +3072,12 @@
       <url hash="e4433f83">2021.naacl-main.197</url>
       <doi>10.18653/v1/2021.naacl-main.197</doi>
       <bibkey>van-der-goot-etal-2021-masked</bibkey>
+      <video href="2021.naacl-main.197.mp4"/>
       <pwccode url="https://github.com/Kaleidophon/deep-significance" additional="true">Kaleidophon/deep-significance</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/xsid">xSID</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/atis">ATIS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.naacl-main.197.mp4"/>
     </paper>
     <paper id="198">
       <title><fixed-case>WEC</fixed-case>: Deriving a Large-scale Cross-document Event Coreference dataset from <fixed-case>W</fixed-case>ikipedia</title>
@@ -3088,10 +3089,10 @@
       <url hash="debe504c">2021.naacl-main.198</url>
       <doi>10.18653/v1/2021.naacl-main.198</doi>
       <bibkey>eirew-etal-2021-wec</bibkey>
+      <video href="2021.naacl-main.198.mp4"/>
       <pwccode url="https://github.com/AlonEirew/extract-wec" additional="true">AlonEirew/extract-wec</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wec-eng">WEC-Eng</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ecb">ECB+</pwcdataset>
-      <video href="2021.naacl-main.198.mp4"/>
     </paper>
     <paper id="199">
       <title>Challenging distributional models with a conceptual network of philosophical terms</title>
@@ -3106,8 +3107,8 @@
       <url hash="ad6a5098">2021.naacl-main.199</url>
       <doi>10.18653/v1/2021.naacl-main.199</doi>
       <bibkey>oortwijn-etal-2021-challenging</bibkey>
-      <pwccode url="https://github.com/yoortwijn/challenging_dms" additional="false">yoortwijn/challenging_dms</pwccode>
       <video href="2021.naacl-main.199.mp4"/>
+      <pwccode url="https://github.com/yoortwijn/challenging_dms" additional="false">yoortwijn/challenging_dms</pwccode>
     </paper>
     <paper id="200">
       <title><fixed-case>KILT</fixed-case>: a Benchmark for Knowledge Intensive Language Tasks</title>
@@ -3129,6 +3130,7 @@
       <url hash="21842ec2">2021.naacl-main.200</url>
       <doi>10.18653/v1/2021.naacl-main.200</doi>
       <bibkey>petroni-etal-2021-kilt</bibkey>
+      <video href="2021.naacl-main.200.mp4"/>
       <pwccode url="https://github.com/facebookresearch/KILT" additional="true">facebookresearch/KILT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/aida-conll-yago">AIDA CoNLL-YAGO</pwcdataset>
@@ -3140,7 +3142,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.naacl-main.200.mp4"/>
     </paper>
     <paper id="201">
       <title>A Survey on Recent Approaches for Natural Language Processing in Low-Resource Scenarios</title>
@@ -3167,8 +3168,8 @@
       <url hash="3e632179">2021.naacl-main.202</url>
       <doi>10.18653/v1/2021.naacl-main.202</doi>
       <bibkey>xu-etal-2021-temporal</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
       <video href="2021.naacl-main.202.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
     </paper>
     <paper id="203">
       <title><fixed-case>UDALM</fixed-case>: Unsupervised Domain Adaptation through Language Modeling</title>
@@ -3180,9 +3181,9 @@
       <url hash="056ceec8">2021.naacl-main.203</url>
       <doi>10.18653/v1/2021.naacl-main.203</doi>
       <bibkey>karouzos-etal-2021-udalm</bibkey>
+      <video href="2021.naacl-main.203.mp4"/>
       <pwccode url="https://github.com/ckarouzos/slp_daptmlm" additional="false">ckarouzos/slp_daptmlm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multi-domain-sentiment-dataset-v2-0">Multi-Domain Sentiment Dataset v2.0</pwcdataset>
-      <video href="2021.naacl-main.203.mp4"/>
     </paper>
     <paper id="204">
       <title>Beyond Black &amp; White: Leveraging Annotator Disagreement via Soft-Label Multi-Task Learning</title>
@@ -3211,8 +3212,8 @@
       <url hash="c5862774">2021.naacl-main.205</url>
       <doi>10.18653/v1/2021.naacl-main.205</doi>
       <bibkey>angell-etal-2021-clustering</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/medmentions">MedMentions</pwcdataset>
       <video href="2021.naacl-main.205.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/medmentions">MedMentions</pwcdataset>
     </paper>
     <paper id="206">
       <title>Variance-reduced First-order Meta-learning for Natural Language Processing Tasks</title>
@@ -3226,8 +3227,8 @@
       <url hash="93468fde">2021.naacl-main.206</url>
       <doi>10.18653/v1/2021.naacl-main.206</doi>
       <bibkey>wang-etal-2021-variance</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
       <video href="2021.naacl-main.206.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
     </paper>
     <paper id="207">
       <title>Diversity-Aware Batch Active Learning for Dependency Parsing</title>
@@ -3240,8 +3241,8 @@
       <url hash="b34ca299">2021.naacl-main.207</url>
       <doi>10.18653/v1/2021.naacl-main.207</doi>
       <bibkey>shi-etal-2021-diversity</bibkey>
-      <pwccode url="https://github.com/tzshi/dpp-al-parsing-naacl21" additional="false">tzshi/dpp-al-parsing-naacl21</pwccode>
       <video href="2021.naacl-main.207.mp4"/>
+      <pwccode url="https://github.com/tzshi/dpp-al-parsing-naacl21" additional="false">tzshi/dpp-al-parsing-naacl21</pwccode>
     </paper>
     <paper id="208">
       <title>How many data points is a prompt worth?</title>
@@ -3254,13 +3255,13 @@
       <award>Outstanding Short Paper</award>
       <doi>10.18653/v1/2021.naacl-main.208</doi>
       <bibkey>le-scao-rush-2021-many</bibkey>
+      <video href="2021.naacl-main.208.mp4"/>
       <pwccode url="https://github.com/TevenLeScao/pet" additional="false">TevenLeScao/pet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multirc">MultiRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wsc">WSC</pwcdataset>
-      <video href="2021.naacl-main.208.mp4"/>
     </paper>
     <paper id="209">
       <title>Can Latent Alignments Improve Autoregressive Machine Translation?</title>
@@ -3283,9 +3284,9 @@
       <url hash="58efe875">2021.naacl-main.210</url>
       <doi>10.18653/v1/2021.naacl-main.210</doi>
       <bibkey>peters-martins-2021-smoothing</bibkey>
+      <video href="2021.naacl-main.210.mp4"/>
       <pwccode url="https://github.com/deep-spin/S7" additional="false">deep-spin/S7</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2016">WMT 2016</pwcdataset>
-      <video href="2021.naacl-main.210.mp4"/>
     </paper>
     <paper id="211">
       <title>Unified Pre-training for Program Understanding and Generation</title>
@@ -3298,10 +3299,10 @@
       <url hash="309df1bd">2021.naacl-main.211</url>
       <doi>10.18653/v1/2021.naacl-main.211</doi>
       <bibkey>ahmad-etal-2021-unified</bibkey>
+      <video href="2021.naacl-main.211.mp4"/>
       <pwccode url="https://github.com/wasiahmad/PLBART" additional="false">wasiahmad/PLBART</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/concode">CONCODE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/codesearchnet">CodeSearchNet</pwcdataset>
-      <video href="2021.naacl-main.211.mp4"/>
     </paper>
     <paper id="212">
       <title>Hyperparameter-free Continuous Learning for Domain Classification in Natural Language Understanding</title>
@@ -3315,8 +3316,8 @@
       <url hash="cb0a6a08">2021.naacl-main.212</url>
       <doi>10.18653/v1/2021.naacl-main.212</doi>
       <bibkey>hua-etal-2021-hyperparameter</bibkey>
-      <pwccode url="https://github.com/tinghua-code/ccfi" additional="false">tinghua-code/ccfi</pwccode>
       <video href="2021.naacl-main.212.mp4"/>
+      <pwccode url="https://github.com/tinghua-code/ccfi" additional="false">tinghua-code/ccfi</pwccode>
     </paper>
     <paper id="213">
       <title>On the Embeddings of Variables in Recurrent Neural Networks for Source Code</title>
@@ -3338,8 +3339,8 @@
       <url hash="c63759b9">2021.naacl-main.214</url>
       <doi>10.18653/v1/2021.naacl-main.214</doi>
       <bibkey>peng-etal-2021-cross</bibkey>
-      <pwccode url="https://github.com/Pzoom522/L1-Refinement" additional="false">Pzoom522/L1-Refinement</pwccode>
       <video href="2021.naacl-main.214.mp4"/>
+      <pwccode url="https://github.com/Pzoom522/L1-Refinement" additional="false">Pzoom522/L1-Refinement</pwccode>
     </paper>
     <paper id="215">
       <title>Semantic Frame Forecast</title>
@@ -3350,12 +3351,12 @@
       <url hash="41d9ea57">2021.naacl-main.215</url>
       <doi>10.18653/v1/2021.naacl-main.215</doi>
       <bibkey>huang-huang-2021-semantic</bibkey>
+      <video href="2021.naacl-main.215.mp4"/>
       <pwccode url="https://github.com/appleternity/FrameForecasting" additional="false">appleternity/FrameForecasting</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coda-19">CODA-19</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
-      <video href="2021.naacl-main.215.mp4"/>
     </paper>
     <paper id="216">
       <title><fixed-case>MUSER</fixed-case>: <fixed-case>MU</fixed-case>ltimodal Stress detection using Emotion Recognition as an Auxiliary Task</title>
@@ -3369,8 +3370,8 @@
       <url hash="dcde72e8">2021.naacl-main.216</url>
       <doi>10.18653/v1/2021.naacl-main.216</doi>
       <bibkey>yao-etal-2021-muser</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/omg-emotion">OMG-Emotion</pwcdataset>
       <video href="2021.naacl-main.216.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/omg-emotion">OMG-Emotion</pwcdataset>
     </paper>
     <paper id="217">
       <title>Learning to Decompose and Organize Complex Tasks</title>
@@ -3384,9 +3385,9 @@
       <url hash="47cfd7dd">2021.naacl-main.217</url>
       <doi>10.18653/v1/2021.naacl-main.217</doi>
       <bibkey>zhang-etal-2021-learning</bibkey>
+      <video href="2021.naacl-main.217.mp4"/>
       <pwccode url="https://github.com/microsoft/mscomplextasks" additional="false">microsoft/mscomplextasks</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihow">WikiHow</pwcdataset>
-      <video href="2021.naacl-main.217.mp4"/>
     </paper>
     <paper id="218">
       <title>Continual Learning for Text Classification with Information Disentanglement Based Regularization</title>
@@ -3400,9 +3401,9 @@
       <url hash="c5ea54f3">2021.naacl-main.218</url>
       <doi>10.18653/v1/2021.naacl-main.218</doi>
       <bibkey>huang-etal-2021-continual</bibkey>
+      <video href="2021.naacl-main.218.mp4"/>
       <pwccode url="https://github.com/GT-SALT/IDBR" additional="false">GT-SALT/IDBR</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
-      <video href="2021.naacl-main.218.mp4"/>
     </paper>
     <paper id="219">
       <title>Learning from Executions for Semantic Parsing</title>
@@ -3414,8 +3415,8 @@
       <url hash="f2a007cc">2021.naacl-main.219</url>
       <doi>10.18653/v1/2021.naacl-main.219</doi>
       <bibkey>wang-etal-2021-learning-executions</bibkey>
-      <pwccode url="https://github.com/berlino/tensor2struct-public" additional="false">berlino/tensor2struct-public</pwccode>
       <video href="2021.naacl-main.219.mp4"/>
+      <pwccode url="https://github.com/berlino/tensor2struct-public" additional="false">berlino/tensor2struct-public</pwccode>
     </paper>
     <paper id="220">
       <title>Learning to Synthesize Data for Semantic Parsing</title>
@@ -3428,8 +3429,8 @@
       <url hash="ec05c9ba">2021.naacl-main.220</url>
       <doi>10.18653/v1/2021.naacl-main.220</doi>
       <bibkey>wang-etal-2021-learning-synthesize</bibkey>
-      <pwccode url="https://github.com/berlino/tensor2struct-public" additional="false">berlino/tensor2struct-public</pwccode>
       <video href="2021.naacl-main.220.mp4"/>
+      <pwccode url="https://github.com/berlino/tensor2struct-public" additional="false">berlino/tensor2struct-public</pwccode>
     </paper>
     <paper id="221">
       <title>Edge: Enriching Knowledge Graph Embeddings with External Text</title>
@@ -3469,12 +3470,12 @@
       <url hash="8d0f4e56">2021.naacl-main.223</url>
       <doi>10.18653/v1/2021.naacl-main.223</doi>
       <bibkey>patel-etal-2021-game</bibkey>
+      <video href="2021.naacl-main.223.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.223.mp4"/>
     </paper>
     <paper id="224">
       <title>Incorporating External Knowledge to Enhance Tabular Reasoning</title>
@@ -3486,11 +3487,11 @@
       <url hash="4651f6f4">2021.naacl-main.224</url>
       <doi>10.18653/v1/2021.naacl-main.224</doi>
       <bibkey>neeraja-etal-2021-incorporating</bibkey>
+      <video href="2021.naacl-main.224.mp4"/>
       <pwccode url="https://github.com/utahnlp/knowledge_infotabs" additional="false">utahnlp/knowledge_infotabs</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/infotabs">InfoTabS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tabfact">TabFact</pwcdataset>
-      <video href="2021.naacl-main.224.mp4"/>
     </paper>
     <paper id="225">
       <title>Compositional Generalization for Neural Semantic Parsing via Span-level Supervised Attention</title>
@@ -3507,8 +3508,8 @@
       <url hash="6ac8b79f">2021.naacl-main.225</url>
       <doi>10.18653/v1/2021.naacl-main.225</doi>
       <bibkey>yin-etal-2021-compositional</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
       <video href="2021.naacl-main.225.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cfq">CFQ</pwcdataset>
     </paper>
     <paper id="226">
       <title>Domain Adaptation for <fixed-case>A</fixed-case>rabic Cross-Domain and Cross-Dialect Sentiment Analysis from Contextualized Word Embedding</title>
@@ -3521,10 +3522,10 @@
       <url hash="f3a203d1">2021.naacl-main.226</url>
       <doi>10.18653/v1/2021.naacl-main.226</doi>
       <bibkey>el-mekki-etal-2021-domain</bibkey>
+      <video href="2021.naacl-main.226.mp4"/>
       <pwccode url="https://github.com/4mekki4/arabic-nlp-da" additional="false">4mekki4/arabic-nlp-da</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/astd">ASTD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tsac">TSAC</pwcdataset>
-      <video href="2021.naacl-main.226.mp4"/>
     </paper>
     <paper id="227">
       <title>Multi-task Learning of Negation and Speculation for Targeted Sentiment Classification</title>
@@ -3537,10 +3538,10 @@
       <attachment type="OptionalSupplementaryCode" hash="08b676cf">2021.naacl-main.227.OptionalSupplementaryCode.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.227</doi>
       <bibkey>moore-barnes-2021-multi</bibkey>
+      <video href="2021.naacl-main.227.mp4"/>
       <pwccode url="https://github.com/jerbarnes/multitask_negation_for_targeted_sentiment" additional="true">jerbarnes/multitask_negation_for_targeted_sentiment</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mams">MAMS</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
-      <video href="2021.naacl-main.227.mp4"/>
     </paper>
     <paper id="228">
       <title>A Disentangled Adversarial Neural Topic Model for Separating Opinions from Plots in User Reviews</title>
@@ -3552,9 +3553,9 @@
       <url hash="7149de0b">2021.naacl-main.228</url>
       <doi>10.18653/v1/2021.naacl-main.228</doi>
       <bibkey>pergola-etal-2021-disentangled</bibkey>
+      <video href="2021.naacl-main.228.mp4"/>
       <pwccode url="https://github.com/gabrer/diatom" additional="false">gabrer/diatom</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
-      <video href="2021.naacl-main.228.mp4"/>
     </paper>
     <paper id="229">
       <title>Graph Ensemble Learning over Multiple Dependency Trees for Aspect-level Sentiment Classification</title>
@@ -3582,9 +3583,9 @@
       <url hash="553ed5de">2021.naacl-main.230</url>
       <doi>10.18653/v1/2021.naacl-main.230</doi>
       <bibkey>turcan-etal-2021-emotion</bibkey>
+      <video href="2021.naacl-main.230.mp4"/>
       <pwccode url="https://github.com/eturcan/emotion-infused" additional="false">eturcan/emotion-infused</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dreaddit">Dreaddit</pwcdataset>
-      <video href="2021.naacl-main.230.mp4"/>
     </paper>
     <paper id="231">
       <title>Aspect-based Sentiment Analysis with Type-aware Graph Convolutional Networks and Layer Ensemble</title>
@@ -3596,9 +3597,9 @@
       <url hash="fba02bd8">2021.naacl-main.231</url>
       <doi>10.18653/v1/2021.naacl-main.231</doi>
       <bibkey>tian-etal-2021-aspect</bibkey>
+      <video href="2021.naacl-main.231.mp4"/>
       <pwccode url="https://github.com/cuhksz-nlp/ASA-TGCN" additional="false">cuhksz-nlp/ASA-TGCN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mams">MAMS</pwcdataset>
-      <video href="2021.naacl-main.231.mp4"/>
     </paper>
     <paper id="232">
       <title>Supertagging-based Parsing with Linear Context-free Rewriting Systems</title>
@@ -3634,10 +3635,10 @@
       <url hash="2f75ff8b">2021.naacl-main.234</url>
       <doi>10.18653/v1/2021.naacl-main.234</doi>
       <bibkey>shi-etal-2021-learning</bibkey>
+      <video href="2021.naacl-main.234.mp4"/>
       <pwccode url="https://github.com/tzshi/nob-naacl21" additional="false">tzshi/nob-naacl21</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qa-srl">QA-SRL</pwcdataset>
-      <video href="2021.naacl-main.234.mp4"/>
     </paper>
     <paper id="235">
       <title>Bot-Adversarial Dialogue for Safe Conversational Agents</title>
@@ -3667,9 +3668,9 @@
       <url hash="4f976089">2021.naacl-main.236</url>
       <doi>10.18653/v1/2021.naacl-main.236</doi>
       <bibkey>babu-etal-2021-non</bibkey>
+      <video href="2021.naacl-main.236.mp4"/>
       <pwccode url="https://github.com/facebookresearch/pytext" additional="false">facebookresearch/pytext</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.naacl-main.236.mp4"/>
     </paper>
     <paper id="237">
       <title>Example-Driven Intent Prediction with Observers</title>
@@ -3680,9 +3681,9 @@
       <url hash="372e6cf5">2021.naacl-main.237</url>
       <doi>10.18653/v1/2021.naacl-main.237</doi>
       <bibkey>mehri-eric-2021-example</bibkey>
+      <video href="2021.naacl-main.237.mp4"/>
       <pwccode url="https://github.com/alexa/dialoglue" additional="false">alexa/dialoglue</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clinc150">CLINC150</pwcdataset>
-      <video href="2021.naacl-main.237.mp4"/>
     </paper>
     <paper id="238">
       <title>Imperfect also Deserves Reward: Multi-Level and Sequential Reward Modeling for Better Dialog Management</title>
@@ -3698,8 +3699,8 @@
       <url hash="acff4505">2021.naacl-main.238</url>
       <doi>10.18653/v1/2021.naacl-main.238</doi>
       <bibkey>hou-etal-2021-imperfect</bibkey>
-      <pwccode url="https://github.com/sherlock1987/SeqReward" additional="false">sherlock1987/SeqReward</pwccode>
       <video href="2021.naacl-main.238.mp4"/>
+      <pwccode url="https://github.com/sherlock1987/SeqReward" additional="false">sherlock1987/SeqReward</pwccode>
     </paper>
     <paper id="239">
       <title>Action-Based Conversations Dataset: A Corpus for Building More In-Depth Task-Oriented Dialogue Systems</title>
@@ -3713,9 +3714,9 @@
       <url hash="66b9ee55">2021.naacl-main.239</url>
       <doi>10.18653/v1/2021.naacl-main.239</doi>
       <bibkey>chen-etal-2021-action</bibkey>
+      <video href="2021.naacl-main.239.mp4"/>
       <pwccode url="https://github.com/asappresearch/abcd" additional="false">asappresearch/abcd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/abcd">ABCD</pwcdataset>
-      <video href="2021.naacl-main.239.mp4"/>
     </paper>
     <paper id="240">
       <title>Controlling Dialogue Generation with Semantic Exemplars</title>
@@ -3728,10 +3729,10 @@
       <url hash="5cfe9929">2021.naacl-main.240</url>
       <doi>10.18653/v1/2021.naacl-main.240</doi>
       <bibkey>gupta-etal-2021-controlling</bibkey>
+      <video href="2021.naacl-main.240.mp4"/>
       <pwccode url="https://github.com/prakharguptaz/EDGE-exemplars" additional="false">prakharguptaz/EDGE-exemplars</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/framenet">FrameNet</pwcdataset>
-      <video href="2021.naacl-main.240.mp4"/>
     </paper>
     <paper id="241">
       <title><fixed-case>COIL</fixed-case>: Revisit Exact Lexical Match in Information Retrieval with Contextualized Inverted List</title>
@@ -3743,8 +3744,8 @@
       <url hash="ddcb37bb">2021.naacl-main.241</url>
       <doi>10.18653/v1/2021.naacl-main.241</doi>
       <bibkey>gao-etal-2021-coil</bibkey>
-      <pwccode url="https://github.com/luyug/COIL" additional="false">luyug/COIL</pwccode>
       <video href="2021.naacl-main.241.mp4"/>
+      <pwccode url="https://github.com/luyug/COIL" additional="false">luyug/COIL</pwccode>
     </paper>
     <paper id="242">
       <title><fixed-case>X</fixed-case>-Class: Text Classification with Extremely Weak Supervision</title>
@@ -3756,9 +3757,9 @@
       <url hash="fef2d803">2021.naacl-main.242</url>
       <doi>10.18653/v1/2021.naacl-main.242</doi>
       <bibkey>wang-etal-2021-x</bibkey>
+      <video href="2021.naacl-main.242.mp4"/>
       <pwccode url="https://github.com/ZihanWangKi/XClass" additional="true">ZihanWangKi/XClass</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
-      <video href="2021.naacl-main.242.mp4"/>
     </paper>
     <paper id="243">
       <title>Fine-tuning Encoders for Improved Monolingual and Zero-shot Polylingual Neural Topic Modeling</title>
@@ -3769,11 +3770,11 @@
       <url hash="af73dfe3">2021.naacl-main.243</url>
       <doi>10.18653/v1/2021.naacl-main.243</doi>
       <bibkey>mueller-dredze-2021-fine</bibkey>
+      <video href="2021.naacl-main.243.mp4"/>
       <pwccode url="https://github.com/aaronmueller/contextualized-topic-models" additional="false">aaronmueller/contextualized-topic-models</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mldoc">MLDoc</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.naacl-main.243.mp4"/>
     </paper>
     <paper id="244">
       <title>Exploring the Relationship Between Algorithm Performance, Vocabulary, and Run-Time in Text Classification</title>
@@ -3785,8 +3786,8 @@
       <url hash="4cec24da">2021.naacl-main.244</url>
       <doi>10.18653/v1/2021.naacl-main.244</doi>
       <bibkey>fearn-etal-2021-exploring</bibkey>
-      <pwccode url="https://github.com/wfearn/preprocessing-paper" additional="false">wfearn/preprocessing-paper</pwccode>
       <video href="2021.naacl-main.244.mp4"/>
+      <pwccode url="https://github.com/wfearn/preprocessing-paper" additional="false">wfearn/preprocessing-paper</pwccode>
     </paper>
     <paper id="245">
       <title>Faithfully Explainable Recommendation via Neural Logic Reasoning</title>
@@ -3800,8 +3801,8 @@
       <url hash="678c78b2">2021.naacl-main.245</url>
       <doi>10.18653/v1/2021.naacl-main.245</doi>
       <bibkey>zhu-etal-2021-faithfully</bibkey>
-      <pwccode url="https://github.com/orcax/LOGER" additional="false">orcax/LOGER</pwccode>
       <video href="2021.naacl-main.245.mp4"/>
+      <pwccode url="https://github.com/orcax/LOGER" additional="false">orcax/LOGER</pwccode>
     </paper>
     <paper id="246">
       <title>You Sound Like Someone Who Watches Drama Movies: Towards Predicting Movie Preferences from Conversational Interactions</title>
@@ -3814,8 +3815,8 @@
       <url hash="ef2f9cae">2021.naacl-main.246</url>
       <doi>10.18653/v1/2021.naacl-main.246</doi>
       <bibkey>volokhin-etal-2021-sound</bibkey>
-      <pwccode url="https://github.com/sergey-volokhin/conversational-movies" additional="false">sergey-volokhin/conversational-movies</pwccode>
       <video href="2021.naacl-main.246.mp4"/>
+      <pwccode url="https://github.com/sergey-volokhin/conversational-movies" additional="false">sergey-volokhin/conversational-movies</pwccode>
     </paper>
     <paper id="247">
       <title>Reading and Acting while Blindfolded: The Need for Semantics in Text Game Agents</title>
@@ -3827,8 +3828,8 @@
       <url hash="b1b7c054">2021.naacl-main.247</url>
       <doi>10.18653/v1/2021.naacl-main.247</doi>
       <bibkey>yao-etal-2021-reading</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/jericho">Jericho</pwcdataset>
       <video href="2021.naacl-main.247.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/jericho">Jericho</pwcdataset>
     </paper>
     <paper id="248">
       <title><fixed-case>SO</fixed-case>r<fixed-case>T</fixed-case>-ing <fixed-case>VQA</fixed-case> Models : Contrastive Gradient Learning for Improved Consistency</title>
@@ -3844,10 +3845,10 @@
       <attachment type="OptionalSupplementaryCode" hash="1925f1d9">2021.naacl-main.248.OptionalSupplementaryCode.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.248</doi>
       <bibkey>dharur-etal-2021-sort</bibkey>
+      <video href="2021.naacl-main.248.mp4"/>
       <pwccode url="https://github.com/sameerdharur/sorting-vqa" additional="false">sameerdharur/sorting-vqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/vqa-hat">VQA-HAT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.naacl-main.248.mp4"/>
     </paper>
     <paper id="249">
       <title>Semi-Supervised Policy Initialization for Playing Games with Language Hints</title>
@@ -3858,8 +3859,8 @@
       <url hash="516fa0e8">2021.naacl-main.249</url>
       <doi>10.18653/v1/2021.naacl-main.249</doi>
       <bibkey>fu-wang-2021-semi</bibkey>
-      <pwccode url="https://github.com/tsujuifu/code_ssi" additional="false">tsujuifu/code_ssi</pwccode>
       <video href="2021.naacl-main.249.mp4"/>
+      <pwccode url="https://github.com/tsujuifu/code_ssi" additional="false">tsujuifu/code_ssi</pwccode>
     </paper>
     <paper id="250">
       <title>Revisiting Document Representations for Large-Scale Zero-Shot Learning</title>
@@ -3870,12 +3871,12 @@
       <url hash="d832c624">2021.naacl-main.250</url>
       <doi>10.18653/v1/2021.naacl-main.250</doi>
       <bibkey>kil-chao-2021-revisiting</bibkey>
+      <video href="2021.naacl-main.250.mp4"/>
       <pwccode url="https://github.com/heendung/vs-zsl" additional="false">heendung/vs-zsl</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/awa-1">AwA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/awa2-1">AwA2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imagenet">ImageNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/apy">aPY</pwcdataset>
-      <video href="2021.naacl-main.250.mp4"/>
     </paper>
     <paper id="251">
       <title>Negative language transfer in learner <fixed-case>E</fixed-case>nglish: A new dataset</title>
@@ -3887,10 +3888,10 @@
       <url hash="99274768">2021.naacl-main.251</url>
       <doi>10.18653/v1/2021.naacl-main.251</doi>
       <bibkey>farias-wanderley-etal-2021-negative</bibkey>
+      <video href="2021.naacl-main.251.mp4"/>
       <pwccode url="https://github.com/EdTeKLA/LanguageTransfer" additional="false">EdTeKLA/LanguageTransfer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fce">FCE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
-      <video href="2021.naacl-main.251.mp4"/>
     </paper>
     <paper id="252">
       <title><fixed-case>S</fixed-case>ent<fixed-case>S</fixed-case>im: Crosslingual Semantic Evaluation of Machine Translation</title>
@@ -3916,10 +3917,10 @@
       <url hash="70a5bd23">2021.naacl-main.253</url>
       <doi>10.18653/v1/2021.naacl-main.253</doi>
       <bibkey>levinboim-etal-2021-quality</bibkey>
+      <video href="2021.naacl-main.253.mp4"/>
       <pwccode url="https://github.com/google-research-datasets/Image-Caption-Quality-Dataset" additional="false">google-research-datasets/Image-Caption-Quality-Dataset</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/image-caption-quality-dataset">Image Caption Quality Dataset</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
-      <video href="2021.naacl-main.253.mp4"/>
     </paper>
     <paper id="254">
       <title><fixed-case>C</fixed-case>a<fixed-case>S</fixed-case>i<fixed-case>N</fixed-case>o: A Corpus of Campsite Negotiation Dialogues for Automatic Negotiation Systems</title>
@@ -3934,9 +3935,9 @@
       <url hash="9232b1e3">2021.naacl-main.254</url>
       <doi>10.18653/v1/2021.naacl-main.254</doi>
       <bibkey>chawla-etal-2021-casino</bibkey>
+      <video href="2021.naacl-main.254.mp4"/>
       <pwccode url="https://github.com/kushalchawla/CaSiNo" additional="false">kushalchawla/CaSiNo</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/casino">CaSiNo</pwcdataset>
-      <video href="2021.naacl-main.254.mp4"/>
     </paper>
     <paper id="255">
       <title>News Headline Grouping as a Challenging <fixed-case>NLU</fixed-case> Task</title>
@@ -3948,9 +3949,9 @@
       <url hash="800ae2c2">2021.naacl-main.255</url>
       <doi>10.18653/v1/2021.naacl-main.255</doi>
       <bibkey>laban-etal-2021-news</bibkey>
+      <video href="2021.naacl-main.255.mp4"/>
       <pwccode url="https://github.com/tingofurro/headline_grouping" additional="false">tingofurro/headline_grouping</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hlgd">HLGD</pwcdataset>
-      <video href="2021.naacl-main.255.mp4"/>
     </paper>
     <paper id="256">
       <title>Olá, Bonjour, Salve! <fixed-case>XFORMAL</fixed-case>: A Benchmark for Multilingual Formality Style Transfer</title>
@@ -3963,9 +3964,9 @@
       <url hash="e5d15c00">2021.naacl-main.256</url>
       <doi>10.18653/v1/2021.naacl-main.256</doi>
       <bibkey>briakou-etal-2021-ola</bibkey>
+      <video href="2021.naacl-main.256.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/opensubtitles">OpenSubtitles</pwcdataset>
-      <video href="2021.naacl-main.256.mp4"/>
     </paper>
     <paper id="257">
       <title>Grouping Words with Semantic Diversity</title>
@@ -3980,8 +3981,8 @@
       <attachment type="OptionalSupplementaryCode" hash="c7dee13a">2021.naacl-main.257.OptionalSupplementaryCode.pdf</attachment>
       <doi>10.18653/v1/2021.naacl-main.257</doi>
       <bibkey>chubarian-etal-2021-grouping</bibkey>
-      <pwccode url="https://github.com/abdulrafae/dg" additional="false">abdulrafae/dg</pwccode>
       <video href="2021.naacl-main.257.mp4"/>
+      <pwccode url="https://github.com/abdulrafae/dg" additional="false">abdulrafae/dg</pwccode>
     </paper>
     <paper id="258">
       <title>Noise Stability Regularization for Improving <fixed-case>BERT</fixed-case> Fine-tuning</title>
@@ -3995,11 +3996,11 @@
       <url hash="db10efc2">2021.naacl-main.258</url>
       <doi>10.18653/v1/2021.naacl-main.258</doi>
       <bibkey>hua-etal-2021-noise</bibkey>
+      <video href="2021.naacl-main.258.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cola">CoLA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.naacl-main.258.mp4"/>
     </paper>
     <paper id="259">
       <title><fixed-case>F</fixed-case>low<fixed-case>P</fixed-case>rior: Learning Expressive Priors for Latent Variable Sentence Models</title>
@@ -4010,9 +4011,9 @@
       <url hash="a8065e6e">2021.naacl-main.259</url>
       <doi>10.18653/v1/2021.naacl-main.259</doi>
       <bibkey>ding-gimpel-2021-flowprior</bibkey>
+      <video href="2021.naacl-main.259.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.naacl-main.259.mp4"/>
     </paper>
     <paper id="260">
       <title><fixed-case>HTCI</fixed-case>nfo<fixed-case>M</fixed-case>ax: A Global Model for Hierarchical Text Classification via Information Maximization</title>
@@ -4026,10 +4027,10 @@
       <url hash="a5b3e9bd">2021.naacl-main.260</url>
       <doi>10.18653/v1/2021.naacl-main.260</doi>
       <bibkey>deng-etal-2021-htcinfomax</bibkey>
+      <video href="2021.naacl-main.260.mp4"/>
       <pwccode url="https://github.com/RingBDStack/HTCInfoMax" additional="false">RingBDStack/HTCInfoMax</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/rcv1">RCV1</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/web-of-science-dataset">WOS</pwcdataset>
-      <video href="2021.naacl-main.260.mp4"/>
     </paper>
     <paper id="261">
       <title>Knowledge Guided Metric Learning for Few-Shot Text Classification</title>
@@ -4044,8 +4045,8 @@
       <url hash="58b0f698">2021.naacl-main.261</url>
       <doi>10.18653/v1/2021.naacl-main.261</doi>
       <bibkey>sui-etal-2021-knowledge</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/nell">NELL</pwcdataset>
       <video href="2021.naacl-main.261.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/nell">NELL</pwcdataset>
     </paper>
     <paper id="262">
       <title>Ensemble of <fixed-case>MRR</fixed-case> and <fixed-case>NDCG</fixed-case> models for Visual Dialog</title>
@@ -4055,9 +4056,9 @@
       <url hash="9b91aca2">2021.naacl-main.262</url>
       <doi>10.18653/v1/2021.naacl-main.262</doi>
       <bibkey>schwartz-2021-ensemble</bibkey>
+      <video href="2021.naacl-main.262.mp4"/>
       <pwccode url="https://github.com/idansc/mrr-ndcg" additional="false">idansc/mrr-ndcg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/visdial">VisDial</pwcdataset>
-      <video href="2021.naacl-main.262.mp4"/>
     </paper>
     <paper id="263">
       <title>Supervised Neural Clustering via Latent Structured Output Learning: Application to Question Intents</title>
@@ -4068,8 +4069,8 @@
       <url hash="019e73cd">2021.naacl-main.263</url>
       <doi>10.18653/v1/2021.naacl-main.263</doi>
       <bibkey>haponchyk-moschitti-2021-supervised</bibkey>
-      <pwccode url="https://github.com/ikernels/intent-qa" additional="false">ikernels/intent-qa</pwccode>
       <video href="2021.naacl-main.263.mp4"/>
+      <pwccode url="https://github.com/ikernels/intent-qa" additional="false">ikernels/intent-qa</pwccode>
     </paper>
     <paper id="264">
       <title><fixed-case>ConVEx</fixed-case>: Data-Efficient and Few-Shot Slot Labeling</title>
@@ -4080,9 +4081,9 @@
       <url hash="554cc226">2021.naacl-main.264</url>
       <doi>10.18653/v1/2021.naacl-main.264</doi>
       <bibkey>henderson-vulic-2021-convex</bibkey>
+      <video href="2021.naacl-main.264.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cc100">CC100</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snips">SNIPS</pwcdataset>
-      <video href="2021.naacl-main.264.mp4"/>
     </paper>
     <paper id="265">
       <title><fixed-case>CREAD</fixed-case>: Combined Resolution of Ellipses and Anaphora in Dialogues</title>
@@ -4098,9 +4099,9 @@
       <url hash="c4f5d834">2021.naacl-main.265</url>
       <doi>10.18653/v1/2021.naacl-main.265</doi>
       <bibkey>tseng-etal-2021-cread</bibkey>
+      <video href="2021.naacl-main.265.mp4"/>
       <pwccode url="https://github.com/apple/ml-cread" additional="false">apple/ml-cread</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mudoco-queryrewrite">MuDoCo_QueryRewrite</pwcdataset>
-      <video href="2021.naacl-main.265.mp4"/>
     </paper>
     <paper id="266">
       <title>Knowledge-Driven Slot Constraints for Goal-Oriented Dialogue Systems</title>
@@ -4112,8 +4113,8 @@
       <url hash="b0ec24f5">2021.naacl-main.266</url>
       <doi>10.18653/v1/2021.naacl-main.266</doi>
       <bibkey>lertvittayakumjorn-etal-2021-knowledge</bibkey>
-      <pwccode url="https://github.com/amazon-research/nlu-slot-constraints" additional="false">amazon-research/nlu-slot-constraints</pwccode>
       <video href="2021.naacl-main.266.mp4"/>
+      <pwccode url="https://github.com/amazon-research/nlu-slot-constraints" additional="false">amazon-research/nlu-slot-constraints</pwccode>
     </paper>
     <paper id="267">
       <title>Clipping Loops for Sample-Efficient Dialogue Policy Optimisation</title>
@@ -4136,8 +4137,8 @@
       <url hash="5a78dc10">2021.naacl-main.268</url>
       <doi>10.18653/v1/2021.naacl-main.268</doi>
       <bibkey>wood-etal-2021-integrating</bibkey>
-      <pwccode url="https://github.com/drevicko/openki" additional="false">drevicko/openki</pwccode>
       <video href="2021.naacl-main.268.mp4"/>
+      <pwccode url="https://github.com/drevicko/openki" additional="false">drevicko/openki</pwccode>
     </paper>
     <paper id="269">
       <title>Noisy-Labeled <fixed-case>NER</fixed-case> with Confidence Estimation</title>
@@ -4153,8 +4154,8 @@
       <url hash="7442f734">2021.naacl-main.269</url>
       <doi>10.18653/v1/2021.naacl-main.269</doi>
       <bibkey>liu-etal-2021-noisy-labeled</bibkey>
-      <pwccode url="https://github.com/liukun95/Noisy-NER-Confidence-Estimation" additional="false">liukun95/Noisy-NER-Confidence-Estimation</pwccode>
       <video href="2021.naacl-main.269.mp4"/>
+      <pwccode url="https://github.com/liukun95/Noisy-NER-Confidence-Estimation" additional="false">liukun95/Noisy-NER-Confidence-Estimation</pwccode>
     </paper>
     <paper id="270">
       <title><fixed-case>TABBIE</fixed-case>: Pretrained Representations of Tabular Data</title>
@@ -4167,8 +4168,8 @@
       <url hash="0bf6dd05">2021.naacl-main.270</url>
       <doi>10.18653/v1/2021.naacl-main.270</doi>
       <bibkey>iida-etal-2021-tabbie</bibkey>
-      <pwccode url="https://github.com/SFIG611/tabbie" additional="false">SFIG611/tabbie</pwccode>
       <video href="2021.naacl-main.270.mp4"/>
+      <pwccode url="https://github.com/SFIG611/tabbie" additional="false">SFIG611/tabbie</pwccode>
     </paper>
     <paper id="271">
       <title>Better Feature Integration for Named Entity Recognition</title>
@@ -4181,9 +4182,9 @@
       <url hash="f81a1a76">2021.naacl-main.271</url>
       <doi>10.18653/v1/2021.naacl-main.271</doi>
       <bibkey>xu-etal-2021-better</bibkey>
+      <video href="2021.naacl-main.271.mp4"/>
       <pwccode url="https://github.com/xuuuluuu/SynLSTM-for-NER" additional="false">xuuuluuu/SynLSTM-for-NER</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ontonotes-5-0">OntoNotes 5.0</pwcdataset>
-      <video href="2021.naacl-main.271.mp4"/>
     </paper>
     <paper id="272">
       <title><fixed-case>ZS</fixed-case>-<fixed-case>BERT</fixed-case>: Towards Zero-Shot Relation Extraction with Attribute Representation Learning</title>
@@ -4194,10 +4195,10 @@
       <url hash="8c77e646">2021.naacl-main.272</url>
       <doi>10.18653/v1/2021.naacl-main.272</doi>
       <bibkey>chen-li-2021-zs</bibkey>
+      <video href="2021.naacl-main.272.mp4"/>
       <pwccode url="https://github.com/dinobby/ZS-BERT" additional="false">dinobby/ZS-BERT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wiki-zsl">Wiki-ZSL</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
-      <video href="2021.naacl-main.272.mp4"/>
     </paper>
     <paper id="273">
       <title>Graph Convolutional Networks for Event Causality Identification with Rich Document-level Structures</title>
@@ -4223,8 +4224,8 @@
       <url hash="2053025f">2021.naacl-main.274</url>
       <doi>10.18653/v1/2021.naacl-main.274</doi>
       <bibkey>lai-etal-2021-context</bibkey>
-      <pwccode url="https://github.com/laituan245/eventcoref" additional="false">laituan245/eventcoref</pwccode>
       <video href="2021.naacl-main.274.mp4"/>
+      <pwccode url="https://github.com/laituan245/eventcoref" additional="false">laituan245/eventcoref</pwccode>
     </paper>
     <paper id="275">
       <title>Multi-Style Transfer with Discriminative Feedback on Disjoint Corpus</title>
@@ -4237,8 +4238,8 @@
       <url hash="b8e9565c">2021.naacl-main.275</url>
       <doi>10.18653/v1/2021.naacl-main.275</doi>
       <bibkey>goyal-etal-2021-multi</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
       <video href="2021.naacl-main.275.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
     </paper>
     <paper id="276">
       <title><fixed-case>FUDGE</fixed-case>: Controlled Text Generation With Future Discriminators</title>
@@ -4249,9 +4250,9 @@
       <url hash="30c68c02">2021.naacl-main.276</url>
       <doi>10.18653/v1/2021.naacl-main.276</doi>
       <bibkey>yang-klein-2021-fudge</bibkey>
+      <video href="2021.naacl-main.276.mp4"/>
       <pwccode url="https://github.com/yangkevin2/naacl-2021-fudge-controlled-generation" additional="true">yangkevin2/naacl-2021-fudge-controlled-generation</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
-      <video href="2021.naacl-main.276.mp4"/>
     </paper>
     <paper id="277">
       <title>Controllable Text Simplification with Explicit Paraphrasing</title>
@@ -4263,8 +4264,8 @@
       <url hash="57919713">2021.naacl-main.277</url>
       <doi>10.18653/v1/2021.naacl-main.277</doi>
       <bibkey>maddela-etal-2021-controllable</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/newsela">Newsela</pwcdataset>
       <video href="2021.naacl-main.277.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/newsela">Newsela</pwcdataset>
     </paper>
     <paper id="278">
       <title>Knowledge Graph Based Synthetic Corpus Generation for Knowledge-Enhanced Language Model Pre-training</title>
@@ -4277,6 +4278,7 @@
       <url hash="cf99df86">2021.naacl-main.278</url>
       <doi>10.18653/v1/2021.naacl-main.278</doi>
       <bibkey>agarwal-etal-2021-knowledge</bibkey>
+      <video href="2021.naacl-main.278.mp4"/>
       <pwccode url="https://github.com/google-research-datasets/KELM-corpus" additional="false">google-research-datasets/KELM-corpus</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/kelm">KELM</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tekgen">TekGen</pwcdataset>
@@ -4284,7 +4286,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
-      <video href="2021.naacl-main.278.mp4"/>
     </paper>
     <paper id="279">
       <title>Choose Your Own Adventure: Paired Suggestions in Collaborative Writing for Evaluating Story Generation Models</title>
@@ -4295,8 +4296,8 @@
       <url hash="de113163">2021.naacl-main.279</url>
       <doi>10.18653/v1/2021.naacl-main.279</doi>
       <bibkey>clark-smith-2021-choose</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
       <video href="2021.naacl-main.279.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
     </paper>
     <paper id="280">
       <title><fixed-case>I</fixed-case>nfo<fixed-case>XLM</fixed-case>: An Information-Theoretic Framework for Cross-Lingual Language Model Pre-Training</title>
@@ -4315,12 +4316,12 @@
       <url hash="c7fdf3ea">2021.naacl-main.280</url>
       <doi>10.18653/v1/2021.naacl-main.280</doi>
       <bibkey>chi-etal-2021-infoxlm</bibkey>
+      <video href="2021.naacl-main.280.mp4"/>
       <pwccode url="" additional="true"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cc100">CC100</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xtreme">XTREME</pwcdataset>
-      <video href="2021.naacl-main.280.mp4"/>
     </paper>
     <paper id="281">
       <title>Context-Interactive Pre-Training for Document Machine Translation</title>
@@ -4361,12 +4362,12 @@
       <url hash="adc0d29d">2021.naacl-main.283</url>
       <doi>10.18653/v1/2021.naacl-main.283</doi>
       <bibkey>mhamdi-etal-2021-x</bibkey>
+      <video href="2021.naacl-main.283.mp4"/>
       <pwccode url="https://github.com/meryemmhamdi1/meta_cross_nlu_qa" additional="false">meryemmhamdi1/meta_cross_nlu_qa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydiqa-goldp">TyDiQA-GoldP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.naacl-main.283.mp4"/>
     </paper>
     <paper id="284">
       <title>Explicit Alignment Objectives for Multilingual Bidirectional Encoders</title>
@@ -4380,10 +4381,10 @@
       <url hash="079276c9">2021.naacl-main.284</url>
       <doi>10.18653/v1/2021.naacl-main.284</doi>
       <bibkey>hu-etal-2021-explicit</bibkey>
+      <video href="2021.naacl-main.284.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tatoeba">Tatoeba</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.naacl-main.284.mp4"/>
     </paper>
     <paper id="285">
       <title>Cross-lingual Cross-modal Pretraining for Multimodal Retrieval</title>
@@ -4395,8 +4396,8 @@
       <url hash="f8eba6c3">2021.naacl-main.285</url>
       <doi>10.18653/v1/2021.naacl-main.285</doi>
       <bibkey>fei-etal-2021-cross</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <video href="2021.naacl-main.285.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
     </paper>
     <paper id="286">
       <title><fixed-case>W</fixed-case>ikipedia Entities as Rendezvous across Languages: Grounding Multilingual Language Models by Predicting <fixed-case>W</fixed-case>ikipedia Hyperlinks</title>
@@ -4408,13 +4409,13 @@
       <url hash="c31f30e1">2021.naacl-main.286</url>
       <doi>10.18653/v1/2021.naacl-main.286</doi>
       <bibkey>calixto-etal-2021-wikipedia</bibkey>
+      <video href="2021.naacl-main.286.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/mlqa">MLQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/paws-x">PAWS-X</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tydi-qa">TyDi QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xl-wic">XL-WiC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xquad">XQuAD</pwcdataset>
-      <video href="2021.naacl-main.286.mp4"/>
     </paper>
     <paper id="287">
       <title>multi<fixed-case>PR</fixed-case>over: Generating Multiple Proofs for Improved Interpretability in Rule Reasoning</title>
@@ -4426,8 +4427,8 @@
       <url hash="fb1552c8">2021.naacl-main.287</url>
       <doi>10.18653/v1/2021.naacl-main.287</doi>
       <bibkey>saha-etal-2021-multiprover</bibkey>
-      <pwccode url="https://github.com/swarnaHub/multiPRover" additional="false">swarnaHub/multiPRover</pwccode>
       <video href="2021.naacl-main.287.mp4"/>
+      <pwccode url="https://github.com/swarnaHub/multiPRover" additional="false">swarnaHub/multiPRover</pwccode>
     </paper>
     <paper id="288">
       <title>Adaptable and Interpretable Neural <fixed-case>M</fixed-case>emory<fixed-case>O</fixed-case>ver Symbolic Knowledge</title>
@@ -4440,9 +4441,9 @@
       <url hash="bb08d5c1">2021.naacl-main.288</url>
       <doi>10.18653/v1/2021.naacl-main.288</doi>
       <bibkey>verga-etal-2021-adaptable</bibkey>
+      <video href="2021.naacl-main.288.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
-      <video href="2021.naacl-main.288.mp4"/>
     </paper>
     <paper id="289">
       <title><fixed-case>CLEVR</fixed-case>_<fixed-case>HYP</fixed-case>: A Challenge Dataset and Baselines for Visual Question Answering with Hypothetical Actions over Images</title>
@@ -4475,9 +4476,9 @@
       <url hash="38d5fdda">2021.naacl-main.290</url>
       <doi>10.18653/v1/2021.naacl-main.290</doi>
       <bibkey>newman-etal-2021-refining</bibkey>
+      <video href="2021.naacl-main.290.mp4"/>
       <pwccode url="https://github.com/bnewm0609/refining-tse" additional="false">bnewm0609/refining-tse</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/blimp">BLiMP</pwcdataset>
-      <video href="2021.naacl-main.290.mp4"/>
     </paper>
     <paper id="291">
       <title>Universal Adversarial Attacks with Natural Triggers for Text Classification</title>
@@ -4490,10 +4491,10 @@
       <url hash="1dc67b3a">2021.naacl-main.291</url>
       <doi>10.18653/v1/2021.naacl-main.291</doi>
       <bibkey>song-etal-2021-universal</bibkey>
+      <video href="2021.naacl-main.291.mp4"/>
       <pwccode url="https://github.com/Hsuan-Tung/universal_attack_natural_trigger" additional="false">Hsuan-Tung/universal_attack_natural_trigger</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.291.mp4"/>
     </paper>
     <paper id="292">
       <title><fixed-case>Q</fixed-case>uadruplet<fixed-case>BERT</fixed-case>: An Efficient Model For Embedding-Based Large-Scale Retrieval</title>
@@ -4507,8 +4508,8 @@
       <url hash="a622b285">2021.naacl-main.292</url>
       <doi>10.18653/v1/2021.naacl-main.292</doi>
       <bibkey>liu-etal-2021-quadrupletbert</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/reqa">ReQA</pwcdataset>
       <video href="2021.naacl-main.292.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/reqa">ReQA</pwcdataset>
     </paper>
     <paper id="293">
       <title>Dynamically Disentangling Social Bias from Task-Oriented Representations with Adversarial Attack</title>
@@ -4522,8 +4523,8 @@
       <url hash="17ba7069">2021.naacl-main.293</url>
       <doi>10.18653/v1/2021.naacl-main.293</doi>
       <bibkey>wang-etal-2021-dynamically</bibkey>
-      <pwccode url="https://github.com/w-lw/debias_adv" additional="false">w-lw/debias_adv</pwccode>
       <video href="2021.naacl-main.293.mp4"/>
+      <pwccode url="https://github.com/w-lw/debias_adv" additional="false">w-lw/debias_adv</pwccode>
     </paper>
     <paper id="294">
       <title>An Empirical Investigation of Bias in the Multimodal Analysis of Financial Earnings Calls</title>
@@ -4535,8 +4536,8 @@
       <url hash="6944fd3e">2021.naacl-main.294</url>
       <doi>10.18653/v1/2021.naacl-main.294</doi>
       <bibkey>sawhney-etal-2021-empirical</bibkey>
-      <pwccode url="https://github.com/midas-research/multimodal-bias-naacl" additional="false">midas-research/multimodal-bias-naacl</pwccode>
       <video href="2021.naacl-main.294.mp4"/>
+      <pwccode url="https://github.com/midas-research/multimodal-bias-naacl" additional="false">midas-research/multimodal-bias-naacl</pwccode>
     </paper>
     <paper id="295">
       <title>Beyond Fair Pay: Ethical Implications of <fixed-case>NLP</fixed-case> Crowdsourcing</title>
@@ -4564,8 +4565,8 @@
       <url hash="195089a3">2021.naacl-main.296</url>
       <doi>10.18653/v1/2021.naacl-main.296</doi>
       <bibkey>jin-etal-2021-transferability</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/hate-speech">Hate Speech</pwcdataset>
       <video href="2021.naacl-main.296.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/hate-speech">Hate Speech</pwcdataset>
     </paper>
     <paper id="297">
       <title>Case Study: Deontological Ethics in <fixed-case>NLP</fixed-case></title>
@@ -4629,8 +4630,8 @@
       <url hash="6622a4a5">2021.naacl-main.301</url>
       <doi>10.18653/v1/2021.naacl-main.301</doi>
       <bibkey>koto-etal-2021-discourse</bibkey>
-      <pwccode url="https://github.com/fajri91/discourse_probing" additional="false">fajri91/discourse_probing</pwccode>
       <video href="2021.naacl-main.301.mp4"/>
+      <pwccode url="https://github.com/fajri91/discourse_probing" additional="false">fajri91/discourse_probing</pwccode>
     </paper>
     <paper id="302">
       <title><fixed-case>U</fixed-case>ni<fixed-case>D</fixed-case>rop: A Simple yet Effective Technique to Improve Transformer without Extra Cost</title>
@@ -4647,8 +4648,8 @@
       <url hash="bb3168c2">2021.naacl-main.302</url>
       <doi>10.18653/v1/2021.naacl-main.302</doi>
       <bibkey>wu-etal-2021-unidrop</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <video href="2021.naacl-main.302.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
     </paper>
     <paper id="303">
       <title>t<fixed-case>WT</fixed-case>–<fixed-case>WT</fixed-case>: A Dataset to Assert the Role of Target Entities for Detecting Stance of Tweets</title>
@@ -4662,8 +4663,8 @@
       <attachment type="OptionalSupplementaryData" hash="7ed02db9">2021.naacl-main.303.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.303</doi>
       <bibkey>kaushal-etal-2021-twt</bibkey>
-      <pwccode url="https://github.com/Ayushk4/bias-stance" additional="true">Ayushk4/bias-stance</pwccode>
       <video href="2021.naacl-main.303.mp4"/>
+      <pwccode url="https://github.com/Ayushk4/bias-stance" additional="true">Ayushk4/bias-stance</pwccode>
     </paper>
     <paper id="304">
       <title>Learning to Learn to be Right for the Right Reasons</title>
@@ -4676,9 +4677,9 @@
       <url hash="1ef72762">2021.naacl-main.304</url>
       <doi>10.18653/v1/2021.naacl-main.304</doi>
       <bibkey>kavumba-etal-2021-learning</bibkey>
+      <video href="2021.naacl-main.304.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/copa">COPA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.naacl-main.304.mp4"/>
     </paper>
     <paper id="305">
       <title>Double Perturbation: On the Robustness of Robustness and Counterfactual Bias Evaluation</title>
@@ -4692,10 +4693,10 @@
       <url hash="0fa9f3a2">2021.naacl-main.305</url>
       <doi>10.18653/v1/2021.naacl-main.305</doi>
       <bibkey>zhang-etal-2021-double</bibkey>
+      <video href="2021.naacl-main.305.mp4"/>
       <pwccode url="https://github.com/chong-z/nlp-second-order-attack" additional="false">chong-z/nlp-second-order-attack</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.305.mp4"/>
     </paper>
     <paper id="306">
       <title>Explaining Neural Network Predictions on Sentence Pairs via Learning Word-Group Masks</title>
@@ -4711,12 +4712,12 @@
       <url hash="ed1975ed">2021.naacl-main.306</url>
       <doi>10.18653/v1/2021.naacl-main.306</doi>
       <bibkey>chen-etal-2021-explaining</bibkey>
+      <video href="2021.naacl-main.306.mp4"/>
       <pwccode url="https://github.com/UVa-NLP/GMASK" additional="false">UVa-NLP/GMASK</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/e-snli">e-SNLI</pwcdataset>
-      <video href="2021.naacl-main.306.mp4"/>
     </paper>
     <paper id="307">
       <title>Almost Free Semantic Draft for Neural Machine Translation</title>
@@ -4727,8 +4728,8 @@
       <url hash="4ae1d23c">2021.naacl-main.307</url>
       <doi>10.18653/v1/2021.naacl-main.307</doi>
       <bibkey>ai-fang-2021-almost</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wmt-2016">WMT 2016</pwcdataset>
       <video href="2021.naacl-main.307.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wmt-2016">WMT 2016</pwcdataset>
     </paper>
     <paper id="308">
       <title>Pruning-then-Expanding Model for Domain Adaptation of Neural Machine Translation</title>
@@ -4740,8 +4741,8 @@
       <url hash="a6a9ddd6">2021.naacl-main.308</url>
       <doi>10.18653/v1/2021.naacl-main.308</doi>
       <bibkey>gu-etal-2021-pruning</bibkey>
-      <pwccode url="https://github.com/ictnlp/PTE-NMT" additional="false">ictnlp/PTE-NMT</pwccode>
       <video href="2021.naacl-main.308.mp4"/>
+      <pwccode url="https://github.com/ictnlp/PTE-NMT" additional="false">ictnlp/PTE-NMT</pwccode>
     </paper>
     <paper id="309">
       <title>Multi-Hop Transformer for Document-Level Machine Translation</title>
@@ -4814,8 +4815,8 @@
       <url hash="3f69e983">2021.naacl-main.313</url>
       <doi>10.18653/v1/2021.naacl-main.313</doi>
       <bibkey>hao-etal-2021-multi</bibkey>
-      <pwccode url="https://github.com/yongchanghao/multi-task-nat" additional="false">yongchanghao/multi-task-nat</pwccode>
       <video href="2021.naacl-main.313.mp4"/>
+      <pwccode url="https://github.com/yongchanghao/multi-task-nat" additional="false">yongchanghao/multi-task-nat</pwccode>
     </paper>
     <paper id="314">
       <title><fixed-case>ER</fixed-case>-<fixed-case>AE</fixed-case>: Differentially Private Text Generation for Authorship Anonymization</title>
@@ -4828,8 +4829,8 @@
       <url hash="c60aee82">2021.naacl-main.314</url>
       <doi>10.18653/v1/2021.naacl-main.314</doi>
       <bibkey>bo-etal-2021-er</bibkey>
-      <pwccode url="https://github.com/McGill-DMaS/AuthorshipAnonymization" additional="true">McGill-DMaS/AuthorshipAnonymization</pwccode>
       <video href="2021.naacl-main.314.mp4"/>
+      <pwccode url="https://github.com/McGill-DMaS/AuthorshipAnonymization" additional="true">McGill-DMaS/AuthorshipAnonymization</pwccode>
     </paper>
     <paper id="315">
       <title>Distantly Supervised Transformers For <fixed-case>E</fixed-case>-Commerce Product <fixed-case>QA</fixed-case></title>
@@ -4883,8 +4884,8 @@
       <url hash="0f9eb258">2021.naacl-main.318</url>
       <doi>10.18653/v1/2021.naacl-main.318</doi>
       <bibkey>tsai-etal-2021-modeling</bibkey>
-      <pwccode url="https://github.com/MiuLab/ICD-Correlation" additional="false">MiuLab/ICD-Correlation</pwccode>
       <video href="2021.naacl-main.318.mp4"/>
+      <pwccode url="https://github.com/MiuLab/ICD-Correlation" additional="false">MiuLab/ICD-Correlation</pwccode>
     </paper>
     <paper id="319">
       <title>Self-Supervised Contrastive Learning for Efficient User Satisfaction Prediction in Conversational Agents</title>
@@ -4948,8 +4949,8 @@
       <url hash="2d477d65">2021.naacl-main.323</url>
       <doi>10.18653/v1/2021.naacl-main.323</doi>
       <bibkey>pryzant-etal-2021-causal</bibkey>
-      <pwccode url="https://github.com/rpryzant/causal-text" additional="false">rpryzant/causal-text</pwccode>
       <video href="2021.naacl-main.323.mp4"/>
+      <pwccode url="https://github.com/rpryzant/causal-text" additional="false">rpryzant/causal-text</pwccode>
     </paper>
     <paper id="324">
       <title>Dynabench: Rethinking Benchmarking in <fixed-case>NLP</fixed-case></title>
@@ -4977,13 +4978,13 @@
       <url hash="a980f5b4">2021.naacl-main.324</url>
       <doi>10.18653/v1/2021.naacl-main.324</doi>
       <bibkey>kiela-etal-2021-dynabench</bibkey>
+      <video href="2021.naacl-main.324.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/anli">ANLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.naacl-main.324.mp4"/>
     </paper>
     <paper id="325">
       <title>Translational <fixed-case>NLP</fixed-case>: A New Paradigm and General Principles for Natural Language Processing Research</title>
@@ -5008,9 +5009,9 @@
       <url hash="41f96046">2021.naacl-main.326</url>
       <doi>10.18653/v1/2021.naacl-main.326</doi>
       <bibkey>xiao-etal-2021-predicting</bibkey>
+      <video href="2021.naacl-main.326.mp4"/>
       <pwccode url="https://github.com/Wendy-Xiao/summ_guided_disco_parser" additional="false">Wendy-Xiao/summ_guided_disco_parser</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
-      <video href="2021.naacl-main.326.mp4"/>
     </paper>
     <paper id="327">
       <title>Probing for Bridging Inference in Transformer Language Models</title>
@@ -5033,9 +5034,9 @@
       <url hash="3c7061ad">2021.naacl-main.328</url>
       <doi>10.18653/v1/2021.naacl-main.328</doi>
       <bibkey>beyer-etal-2021-incoherence</bibkey>
+      <video href="2021.naacl-main.328.mp4"/>
       <pwccode url="https://github.com/AnneBeyer/coherencegym" additional="false">AnneBeyer/coherencegym</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/rocstories">ROCStories</pwcdataset>
-      <video href="2021.naacl-main.328.mp4"/>
     </paper>
     <paper id="329">
       <title>Stay Together: A System for Single and Split-antecedent Anaphora Resolution</title>
@@ -5048,8 +5049,8 @@
       <url hash="54e69f35">2021.naacl-main.329</url>
       <doi>10.18653/v1/2021.naacl-main.329</doi>
       <bibkey>yu-etal-2021-stay</bibkey>
-      <pwccode url="https://github.com/juntaoy/dali-full-anaphora" additional="false">juntaoy/dali-full-anaphora</pwccode>
       <video href="2021.naacl-main.329.mp4"/>
+      <pwccode url="https://github.com/juntaoy/dali-full-anaphora" additional="false">juntaoy/dali-full-anaphora</pwccode>
     </paper>
     <paper id="330">
       <title>Redefining Absent Keyphrases and their Effect on Retrieval Effectiveness</title>
@@ -5060,8 +5061,8 @@
       <url hash="bf8ad20e">2021.naacl-main.330</url>
       <doi>10.18653/v1/2021.naacl-main.330</doi>
       <bibkey>boudin-gallina-2021-redefining</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
       <video href="2021.naacl-main.330.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
     </paper>
     <paper id="331">
       <title><fixed-case>C</fixed-case>o<fixed-case>RT</fixed-case>: Complementary Rankings from Transformers</title>
@@ -5072,9 +5073,9 @@
       <url hash="65b0ff95">2021.naacl-main.331</url>
       <doi>10.18653/v1/2021.naacl-main.331</doi>
       <bibkey>wrzalik-krechel-2021-cort</bibkey>
+      <video href="2021.naacl-main.331.mp4"/>
       <pwccode url="https://github.com/lavis-nlp/CoRT" additional="false">lavis-nlp/CoRT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
-      <video href="2021.naacl-main.331.mp4"/>
     </paper>
     <paper id="332">
       <title>Multi-source Neural Topic Modeling in Multi-view Embedding Spaces</title>
@@ -5086,8 +5087,8 @@
       <url hash="993d4961">2021.naacl-main.332</url>
       <doi>10.18653/v1/2021.naacl-main.332</doi>
       <bibkey>gupta-etal-2021-multi</bibkey>
-      <pwccode url="https://github.com/YatinChaudhary/Multi-view-Multi-source-Topic-Modeling" additional="false">YatinChaudhary/Multi-view-Multi-source-Topic-Modeling</pwccode>
       <video href="2021.naacl-main.332.mp4"/>
+      <pwccode url="https://github.com/YatinChaudhary/Multi-view-Multi-source-Topic-Modeling" additional="false">YatinChaudhary/Multi-view-Multi-source-Topic-Modeling</pwccode>
     </paper>
     <paper id="333">
       <title>Inductive Topic Variational Graph Auto-Encoder for Text Classification</title>
@@ -5115,11 +5116,11 @@
       <url hash="531398d7">2021.naacl-main.334</url>
       <doi>10.18653/v1/2021.naacl-main.334</doi>
       <bibkey>liu-etal-2021-self</bibkey>
+      <video href="2021.naacl-main.334.mp4"/>
       <pwccode url="https://github.com/cambridgeltl/sapbert" additional="false">cambridgeltl/sapbert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/blue">BLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/cometa">COMETA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/medmentions">MedMentions</pwcdataset>
-      <video href="2021.naacl-main.334.mp4"/>
     </paper>
     <paper id="335">
       <title><fixed-case>T</fixed-case>axo<fixed-case>C</fixed-case>lass: Hierarchical Multi-Label Text Classification Using Only Class Names</title>
@@ -5147,9 +5148,9 @@
       <url hash="ea2cd7b4">2021.naacl-main.336</url>
       <doi>10.18653/v1/2021.naacl-main.336</doi>
       <bibkey>chakrabarty-etal-2021-mermaid</bibkey>
+      <video href="2021.naacl-main.336.mp4"/>
       <pwccode url="https://github.com/tuhinjubcse/MetaphorGenNAACL2021" additional="false">tuhinjubcse/MetaphorGenNAACL2021</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
-      <video href="2021.naacl-main.336.mp4"/>
     </paper>
     <paper id="337">
       <title>On Learning Text Style Transfer with Direct Rewards</title>
@@ -5161,9 +5162,9 @@
       <url hash="77107d4f">2021.naacl-main.337</url>
       <doi>10.18653/v1/2021.naacl-main.337</doi>
       <bibkey>liu-etal-2021-learning</bibkey>
+      <video href="2021.naacl-main.337.mp4"/>
       <pwccode url="https://github.com/yixinL7/Direct-Style-Transfer" additional="false">yixinL7/Direct-Style-Transfer</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
-      <video href="2021.naacl-main.337.mp4"/>
     </paper>
     <paper id="338">
       <title>Focused Attention Improves Document-Grounded Generation</title>
@@ -5177,9 +5178,9 @@
       <url hash="d554e4fc">2021.naacl-main.338</url>
       <doi>10.18653/v1/2021.naacl-main.338</doi>
       <bibkey>prabhumoye-etal-2021-focused</bibkey>
+      <video href="2021.naacl-main.338.mp4"/>
       <pwccode url="https://github.com/shrimai/Focused-Attention-Improves-Document-Grounded-Generation" additional="false">shrimai/Focused-Attention-Improves-Document-Grounded-Generation</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
-      <video href="2021.naacl-main.338.mp4"/>
     </paper>
     <paper id="339">
       <title><fixed-case>N</fixed-case>euro<fixed-case>L</fixed-case>ogic Decoding: (Un)supervised Neural Text Generation with Predicate Logic Constraints</title>
@@ -5194,8 +5195,8 @@
       <url hash="9227069d">2021.naacl-main.339</url>
       <doi>10.18653/v1/2021.naacl-main.339</doi>
       <bibkey>lu-etal-2021-neurologic</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/commongen">CommonGen</pwcdataset>
       <video href="2021.naacl-main.339.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/commongen">CommonGen</pwcdataset>
     </paper>
     <paper id="340">
       <title>Ask what’s missing and what’s useful: Improving Clarification Question Generation using Global Knowledge</title>
@@ -5208,8 +5209,8 @@
       <url hash="c019b5d5">2021.naacl-main.340</url>
       <doi>10.18653/v1/2021.naacl-main.340</doi>
       <bibkey>majumder-etal-2021-ask</bibkey>
-      <pwccode url="https://github.com/microsoft/clarification-qgen-globalinfo" additional="false">microsoft/clarification-qgen-globalinfo</pwccode>
       <video href="2021.naacl-main.340.mp4"/>
+      <pwccode url="https://github.com/microsoft/clarification-qgen-globalinfo" additional="false">microsoft/clarification-qgen-globalinfo</pwccode>
     </paper>
     <paper id="341">
       <title>Progressive Generation of Long Text with Pretrained Language Models</title>
@@ -5223,9 +5224,9 @@
       <url hash="cab8615e">2021.naacl-main.341</url>
       <doi>10.18653/v1/2021.naacl-main.341</doi>
       <bibkey>tan-etal-2021-progressive</bibkey>
+      <video href="2021.naacl-main.341.mp4"/>
       <pwccode url="https://github.com/tanyuqian/progressive-generation" additional="false">tanyuqian/progressive-generation</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/writingprompts">WritingPrompts</pwcdataset>
-      <video href="2021.naacl-main.341.mp4"/>
     </paper>
     <paper id="342">
       <title><fixed-case>SOCCER</fixed-case>: An Information-Sparse Discourse State Tracking Collection in the Sports Commentary Domain</title>
@@ -5236,9 +5237,9 @@
       <url hash="dc9258a0">2021.naacl-main.342</url>
       <doi>10.18653/v1/2021.naacl-main.342</doi>
       <bibkey>zhang-eickhoff-2021-soccer</bibkey>
+      <video href="2021.naacl-main.342.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/open-pi">Open PI</pwcdataset>
-      <video href="2021.naacl-main.342.mp4"/>
     </paper>
     <paper id="343">
       <title>Plot-guided Adversarial Example Construction for Evaluating Open-domain Story Generation</title>
@@ -5255,8 +5256,8 @@
       <revision id="2" href="2021.naacl-main.343v2" hash="9436c6d1" date="2021-05-25">Updated a citation.</revision>
       <doi>10.18653/v1/2021.naacl-main.343</doi>
       <bibkey>ghazarian-etal-2021-plot</bibkey>
-      <pwccode url="https://github.com/PlusLabNLP/Plot-guided-Coherence-Evaluation" additional="false">PlusLabNLP/Plot-guided-Coherence-Evaluation</pwccode>
       <video href="2021.naacl-main.343.mp4"/>
+      <pwccode url="https://github.com/PlusLabNLP/Plot-guided-Coherence-Evaluation" additional="false">PlusLabNLP/Plot-guided-Coherence-Evaluation</pwccode>
     </paper>
     <paper id="344">
       <title><fixed-case>M</fixed-case>ulti<fixed-case>O</fixed-case>p<fixed-case>E</fixed-case>d: A Corpus of Multi-Perspective News Editorials</title>
@@ -5271,9 +5272,9 @@
       <attachment type="OptionalSupplementaryCode" hash="c5673a7e">2021.naacl-main.344.OptionalSupplementaryCode.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.344</doi>
       <bibkey>liu-etal-2021-multioped</bibkey>
+      <video href="2021.naacl-main.344.mp4"/>
       <pwccode url="https://github.com/CogComp/MultiOpEd" additional="false">CogComp/MultiOpEd</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multioped">MultiOpEd</pwcdataset>
-      <video href="2021.naacl-main.344.mp4"/>
     </paper>
     <paper id="345">
       <title>Swords: A Benchmark for Lexical Substitution with Improved Data Coverage and Quality</title>
@@ -5287,9 +5288,9 @@
       <url hash="14770f81">2021.naacl-main.345</url>
       <doi>10.18653/v1/2021.naacl-main.345</doi>
       <bibkey>lee-etal-2021-swords</bibkey>
+      <video href="2021.naacl-main.345.mp4"/>
       <pwccode url="https://github.com/p-lambda/swords" additional="false">p-lambda/swords</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/swords">Swords</pwcdataset>
-      <video href="2021.naacl-main.345.mp4"/>
     </paper>
     <paper id="346">
       <title>“<fixed-case>I</fixed-case>’m Not Mad”: Commonsense Implications of Negation and Contradiction</title>
@@ -5302,9 +5303,9 @@
       <url hash="31b513c7">2021.naacl-main.346</url>
       <doi>10.18653/v1/2021.naacl-main.346</doi>
       <bibkey>jiang-etal-2021-im</bibkey>
+      <video href="2021.naacl-main.346.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/atomic">ATOMIC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
-      <video href="2021.naacl-main.346.mp4"/>
     </paper>
     <paper id="347">
       <title>Identifying Medical Self-Disclosure in Online Communities</title>
@@ -5343,8 +5344,8 @@
       <url hash="e918e632">2021.naacl-main.349</url>
       <doi>10.18653/v1/2021.naacl-main.349</doi>
       <bibkey>pimentel-etal-2021-finding</bibkey>
-      <pwccode url="https://github.com/tpimentelms/form-meaning-associations" additional="true">tpimentelms/form-meaning-associations</pwccode>
       <video href="2021.naacl-main.349.mp4"/>
+      <pwccode url="https://github.com/tpimentelms/form-meaning-associations" additional="true">tpimentelms/form-meaning-associations</pwccode>
     </paper>
     <paper id="350">
       <title>How (Non-)Optimal is the Lexicon?</title>
@@ -5397,8 +5398,8 @@
       <url hash="9e8f4679">2021.naacl-main.353</url>
       <doi>10.18653/v1/2021.naacl-main.353</doi>
       <bibkey>meloni-etal-2021-ab</bibkey>
-      <pwccode url="https://github.com/shauli-ravfogel/Latin-Reconstruction-NAACL" additional="false">shauli-ravfogel/Latin-Reconstruction-NAACL</pwccode>
       <video href="2021.naacl-main.353.mp4"/>
+      <pwccode url="https://github.com/shauli-ravfogel/Latin-Reconstruction-NAACL" additional="false">shauli-ravfogel/Latin-Reconstruction-NAACL</pwccode>
     </paper>
     <paper id="354">
       <title>On Biasing Transformer Attention Towards Monotonicity</title>
@@ -5429,10 +5430,10 @@
       <url hash="d9e9c175">2021.naacl-main.355</url>
       <doi>10.18653/v1/2021.naacl-main.355</doi>
       <bibkey>hope-etal-2021-extracting</bibkey>
+      <video href="2021.naacl-main.355.mp4"/>
       <pwccode url="https://github.com/dwadden/dygiepp" additional="true">dwadden/dygiepp</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cord-19">CORD-19</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scierc">SciERC</pwcdataset>
-      <video href="2021.naacl-main.355.mp4"/>
     </paper>
     <paper id="356">
       <title>Constrained Multi-Task Learning for Event Coreference Resolution</title>
@@ -5457,8 +5458,8 @@
       <url hash="c1b9bbd9">2021.naacl-main.357</url>
       <doi>10.18653/v1/2021.naacl-main.357</doi>
       <bibkey>v-ganesan-etal-2021-empirical</bibkey>
-      <pwccode url="https://github.com/adithya8/ContextualEmbeddingDR" additional="false">adithya8/ContextualEmbeddingDR</pwccode>
       <video href="2021.naacl-main.357.mp4"/>
+      <pwccode url="https://github.com/adithya8/ContextualEmbeddingDR" additional="false">adithya8/ContextualEmbeddingDR</pwccode>
     </paper>
     <paper id="358">
       <title>Leveraging Deep Representations of Radiology Reports in Survival Analysis for Predicting Heart Failure Patient Mortality</title>
@@ -5472,10 +5473,10 @@
       <url hash="39b07bce">2021.naacl-main.358</url>
       <doi>10.18653/v1/2021.naacl-main.358</doi>
       <bibkey>lee-etal-2021-leveraging</bibkey>
+      <video href="2021.naacl-main.358.mp4"/>
       <pwccode url="https://github.com/bionlplab/heart_failure_mortality" additional="false">bionlplab/heart_failure_mortality</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/blue">BLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/chexpert">CheXpert</pwcdataset>
-      <video href="2021.naacl-main.358.mp4"/>
     </paper>
     <paper id="359">
       <title>On the Use of Context for Predicting Citation Worthiness of Sentences in Scholarly Articles</title>
@@ -5502,8 +5503,8 @@
       <url hash="002c4f13">2021.naacl-main.360</url>
       <doi>10.18653/v1/2021.naacl-main.360</doi>
       <bibkey>mithun-etal-2021-data</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/figer">FIGER</pwcdataset>
       <video href="2021.naacl-main.360.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/figer">FIGER</pwcdataset>
     </paper>
     <paper id="361">
       <title>Adapting Coreference Resolution for Processing Violent Death Narratives</title>
@@ -5518,10 +5519,10 @@
       <url hash="00bf1116">2021.naacl-main.361</url>
       <doi>10.18653/v1/2021.naacl-main.361</doi>
       <bibkey>uppunda-etal-2021-adapting</bibkey>
+      <video href="2021.naacl-main.361.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/gap-coreference-dataset">GAP Coreference Dataset</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/gicoref">GICoref</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/map">MAP</pwcdataset>
-      <video href="2021.naacl-main.361.mp4"/>
     </paper>
     <paper id="362">
       <title>Time-Stamped Language Model: Teaching Language Models to Understand The Flow of Events</title>
@@ -5532,9 +5533,9 @@
       <url hash="f75557eb">2021.naacl-main.362</url>
       <doi>10.18653/v1/2021.naacl-main.362</doi>
       <bibkey>rajaby-faghihi-kordjamshidi-2021-time</bibkey>
+      <video href="2021.naacl-main.362.mp4"/>
       <pwccode url="https://github.com/HLR/TSLM" additional="false">HLR/TSLM</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/propara">ProPara</pwcdataset>
-      <video href="2021.naacl-main.362.mp4"/>
     </paper>
     <paper id="363">
       <title>If You Want to Go Far Go Together: Unsupervised Joint Candidate Evidence Retrieval for Multi-hop Question Answering</title>
@@ -5546,10 +5547,10 @@
       <url hash="2c07eff9">2021.naacl-main.363</url>
       <doi>10.18653/v1/2021.naacl-main.363</doi>
       <bibkey>yadav-etal-2021-want</bibkey>
+      <video href="2021.naacl-main.363.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/multirc">MultiRC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qasc">QASC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.naacl-main.363.mp4"/>
     </paper>
     <paper id="364">
       <title><fixed-case>SPARTQA</fixed-case>: A Textual Question Answering Benchmark for Spatial Reasoning</title>
@@ -5562,12 +5563,11 @@
       <url hash="474d2b28">2021.naacl-main.364</url>
       <doi>10.18653/v1/2021.naacl-main.364</doi>
       <bibkey>mirzaee-etal-2021-spartqa</bibkey>
+      <video href="2021.naacl-main.364.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/spartqa-1">SPARTQA -</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/boolq">BoolQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/nlvr">NLVR</pwcdataset>
-      <pwcdataset url="https://paperswithcode.com/dataset/babi-1">bAbI</pwcdataset>
-      <video href="2021.naacl-main.364.mp4"/>
     </paper>
     <paper id="365">
       <title>A Dataset of Information-Seeking Questions and Answers Anchored in Research Papers</title>
@@ -5582,10 +5582,10 @@
       <url hash="fe6e9970">2021.naacl-main.365</url>
       <doi>10.18653/v1/2021.naacl-main.365</doi>
       <bibkey>dasigi-etal-2021-dataset</bibkey>
+      <video href="2021.naacl-main.365.mp4"/>
       <pwccode url="https://github.com/allenai/qasper-led-baseline" additional="false">allenai/qasper-led-baseline</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/qasper">QASPER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/s2orc">S2ORC</pwcdataset>
-      <video href="2021.naacl-main.365.mp4"/>
     </paper>
     <paper id="366">
       <title>Differentiable Open-Ended Commonsense Reasoning</title>
@@ -5600,12 +5600,12 @@
       <url hash="a3985c3c">2021.naacl-main.366</url>
       <doi>10.18653/v1/2021.naacl-main.366</doi>
       <bibkey>lin-etal-2021-differentiable</bibkey>
+      <video href="2021.naacl-main.366.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/arc">ARC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/genericskb">GenericsKB</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qasc">QASC</pwcdataset>
-      <video href="2021.naacl-main.366.mp4"/>
     </paper>
     <paper id="367">
       <title>Does Structure Matter? Encoding Documents for Machine Reading Comprehension</title>
@@ -5620,10 +5620,10 @@
       <url hash="23216f3b">2021.naacl-main.367</url>
       <doi>10.18653/v1/2021.naacl-main.367</doi>
       <bibkey>wan-etal-2021-structure</bibkey>
+      <video href="2021.naacl-main.367.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/doc2dial-1">Doc2Dial</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/doc2dial">doc2dial</pwcdataset>
-      <video href="2021.naacl-main.367.mp4"/>
     </paper>
     <paper id="368">
       <title>Multi-Step Reasoning Over Unstructured Text with Beam Dense Retrieval</title>
@@ -5636,8 +5636,8 @@
       <url hash="7eb4993c">2021.naacl-main.368</url>
       <doi>10.18653/v1/2021.naacl-main.368</doi>
       <bibkey>zhao-etal-2021-multi-step</bibkey>
-      <pwccode url="https://github.com/henryzhao5852/BeamDR" additional="false">henryzhao5852/BeamDR</pwccode>
       <video href="2021.naacl-main.368.mp4"/>
+      <pwccode url="https://github.com/henryzhao5852/BeamDR" additional="false">henryzhao5852/BeamDR</pwccode>
     </paper>
     <paper id="369">
       <title>Scalable and Interpretable Semantic Change Detection</title>
@@ -5649,8 +5649,8 @@
       <url hash="a54bb7be">2021.naacl-main.369</url>
       <doi>10.18653/v1/2021.naacl-main.369</doi>
       <bibkey>montariol-etal-2021-scalable</bibkey>
-      <pwccode url="https://github.com/matejmartinc/scalable_semantic_shift" additional="false">matejmartinc/scalable_semantic_shift</pwccode>
       <video href="2021.naacl-main.369.mp4"/>
+      <pwccode url="https://github.com/matejmartinc/scalable_semantic_shift" additional="false">matejmartinc/scalable_semantic_shift</pwccode>
     </paper>
     <paper id="370">
       <title>Scalar Adjective Identification and Multilingual Ranking</title>
@@ -5673,9 +5673,9 @@
       <url hash="c245a034">2021.naacl-main.371</url>
       <doi>10.18653/v1/2021.naacl-main.371</doi>
       <bibkey>barba-etal-2021-esc</bibkey>
+      <video href="2021.naacl-main.371.mp4"/>
       <pwccode url="https://github.com/SapienzaNLP/esc" additional="false">SapienzaNLP/esc</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/word-sense-disambiguation-a-unified">Word Sense Disambiguation: a Unified Evaluation Framework and Empirical Comparison</pwcdataset>
-      <video href="2021.naacl-main.371.mp4"/>
     </paper>
     <paper id="372">
       <title>Recent advances in neural metaphor processing: A linguistic, cognitive and social perspective</title>
@@ -5724,8 +5724,8 @@
       <url hash="f0e16946">2021.naacl-main.375</url>
       <doi>10.18653/v1/2021.naacl-main.375</doi>
       <bibkey>huang-etal-2021-seq2emo</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/goemotions">GoEmotions</pwcdataset>
       <video href="2021.naacl-main.375.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/goemotions">GoEmotions</pwcdataset>
     </paper>
     <paper id="376">
       <title>Knowledge Enhanced Masked Language Model for Stance Detection</title>
@@ -5736,8 +5736,8 @@
       <url hash="8e4dfafe">2021.naacl-main.376</url>
       <doi>10.18653/v1/2021.naacl-main.376</doi>
       <bibkey>kawintiranon-singh-2021-knowledge</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/twitter-stance-election-2020">Twitter Stance Election 2020</pwcdataset>
       <video href="2021.naacl-main.376.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/twitter-stance-election-2020">Twitter Stance Election 2020</pwcdataset>
     </paper>
     <paper id="377">
       <title>Learning Paralinguistic Features from Audiobooks through Style Voice Conversion</title>
@@ -5749,8 +5749,8 @@
       <url hash="25df5c7a">2021.naacl-main.377</url>
       <doi>10.18653/v1/2021.naacl-main.377</doi>
       <bibkey>aldeneh-etal-2021-learning</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <video href="2021.naacl-main.377.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
     </paper>
     <paper id="378">
       <title>Adapting <fixed-case>BERT</fixed-case> for Continual Learning of a Sequence of Aspect Sentiment Classification Tasks</title>
@@ -5762,11 +5762,11 @@
       <url hash="f17af448">2021.naacl-main.378</url>
       <doi>10.18653/v1/2021.naacl-main.378</doi>
       <bibkey>ke-etal-2021-adapting</bibkey>
+      <video href="2021.naacl-main.378.mp4"/>
       <pwccode url="https://github.com/zixuanke/pycontinual" additional="false">zixuanke/pycontinual</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/asc-til-19-tasks">ASC (TIL, 19 tasks)</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/20newsgroup-10-tasks">20Newsgroup (10 tasks)</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dsc-10-tasks">DSC (10 tasks)</pwcdataset>
-      <video href="2021.naacl-main.378.mp4"/>
     </paper>
     <paper id="379">
       <title>Adversarial Learning for Zero-Shot Stance Detection on Social Media</title>
@@ -5778,8 +5778,8 @@
       <url hash="a1d5a465">2021.naacl-main.379</url>
       <doi>10.18653/v1/2021.naacl-main.379</doi>
       <bibkey>allaway-etal-2021-adversarial</bibkey>
-      <pwccode url="https://github.com/MalavikaSrikanth16/adversarial-learning-for-stance" additional="false">MalavikaSrikanth16/adversarial-learning-for-stance</pwccode>
       <video href="2021.naacl-main.379.mp4"/>
+      <pwccode url="https://github.com/MalavikaSrikanth16/adversarial-learning-for-stance" additional="false">MalavikaSrikanth16/adversarial-learning-for-stance</pwccode>
     </paper>
     <paper id="380">
       <title>Efficiently Summarizing Text and Graph Encodings of Multi-Document Clusters</title>
@@ -5793,9 +5793,9 @@
       <url hash="10ad2f63">2021.naacl-main.380</url>
       <doi>10.18653/v1/2021.naacl-main.380</doi>
       <bibkey>pasunuru-etal-2021-efficiently</bibkey>
+      <video href="2021.naacl-main.380.mp4"/>
       <pwccode url="https://github.com/amazon-research/bartgraphsumm" additional="false">amazon-research/bartgraphsumm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisum">WikiSum</pwcdataset>
-      <video href="2021.naacl-main.380.mp4"/>
     </paper>
     <paper id="381">
       <title>Enriching Transformers with Structured Tensor-Product Representations for Abstractive Summarization</title>
@@ -5814,10 +5814,10 @@
       <url hash="f06a0ec4">2021.naacl-main.381</url>
       <doi>10.18653/v1/2021.naacl-main.381</doi>
       <bibkey>jiang-etal-2021-enriching</bibkey>
+      <video href="2021.naacl-main.381.mp4"/>
       <pwccode url="https://github.com/jiangycTarheel/TPT-Summ" additional="false">jiangycTarheel/TPT-Summ</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihow">WikiHow</pwcdataset>
-      <video href="2021.naacl-main.381.mp4"/>
     </paper>
     <paper id="382">
       <title>What’s in a Summary? Laying the Groundwork for Advances in Hospital-Course Summarization</title>
@@ -5831,8 +5831,8 @@
       <url hash="9a4381c2">2021.naacl-main.382</url>
       <doi>10.18653/v1/2021.naacl-main.382</doi>
       <bibkey>adams-etal-2021-whats</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <video href="2021.naacl-main.382.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
     </paper>
     <paper id="383">
       <title>Understanding Factuality in Abstractive Summarization with <fixed-case>FRANK</fixed-case>: A Benchmark for Factuality Metrics</title>
@@ -5844,8 +5844,8 @@
       <url hash="0cb57592">2021.naacl-main.383</url>
       <doi>10.18653/v1/2021.naacl-main.383</doi>
       <bibkey>pagnoni-etal-2021-understanding</bibkey>
-      <pwccode url="https://github.com/artidoro/frank" additional="false">artidoro/frank</pwccode>
       <video href="2021.naacl-main.383.mp4"/>
+      <pwccode url="https://github.com/artidoro/frank" additional="false">artidoro/frank</pwccode>
     </paper>
     <paper id="384">
       <title><fixed-case>GS</fixed-case>um: A General Framework for Guided Neural Abstractive Summarization</title>
@@ -5859,10 +5859,10 @@
       <url hash="8bfb84c8">2021.naacl-main.384</url>
       <doi>10.18653/v1/2021.naacl-main.384</doi>
       <bibkey>dou-etal-2021-gsum</bibkey>
+      <video href="2021.naacl-main.384.mp4"/>
       <pwccode url="https://github.com/neulab/guided_summarization" additional="false">neulab/guided_summarization</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihow">WikiHow</pwcdataset>
-      <video href="2021.naacl-main.384.mp4"/>
     </paper>
     <paper id="385">
       <title>What Will it Take to Fix Benchmarking in Natural Language Understanding?</title>
@@ -5873,11 +5873,11 @@
       <url hash="759a9457">2021.naacl-main.385</url>
       <doi>10.18653/v1/2021.naacl-main.385</doi>
       <bibkey>bowman-dahl-2021-will</bibkey>
+      <video href="2021.naacl-main.385.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.naacl-main.385.mp4"/>
     </paper>
     <paper id="386">
       <title><fixed-case>T</fixed-case>uring<fixed-case>A</fixed-case>dvice: A Generative and Dynamic Evaluation of Language Use</title>
@@ -5892,11 +5892,11 @@
       <url hash="1b9451ed">2021.naacl-main.386</url>
       <doi>10.18653/v1/2021.naacl-main.386</doi>
       <bibkey>zellers-etal-2021-turingadvice</bibkey>
+      <video href="2021.naacl-main.386.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hellaswag">HellaSwag</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/swag">SWAG</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
-      <video href="2021.naacl-main.386.mp4"/>
     </paper>
     <paper id="387">
       <title>Multitask Learning for Emotionally Analyzing Sexual Abuse Disclosures</title>
@@ -5910,8 +5910,8 @@
       <url hash="d35db9e4">2021.naacl-main.387</url>
       <doi>10.18653/v1/2021.naacl-main.387</doi>
       <bibkey>sawhney-etal-2021-multitask</bibkey>
-      <pwccode url="https://github.com/midas-research/metoo-mtl-naacl" additional="false">midas-research/metoo-mtl-naacl</pwccode>
       <video href="2021.naacl-main.387.mp4"/>
+      <pwccode url="https://github.com/midas-research/metoo-mtl-naacl" additional="false">midas-research/metoo-mtl-naacl</pwccode>
     </paper>
     <paper id="388">
       <title>Self Promotion in <fixed-case>US</fixed-case> Congressional Tweets</title>
@@ -5923,8 +5923,8 @@
       <url hash="b851d9d2">2021.naacl-main.388</url>
       <doi>10.18653/v1/2021.naacl-main.388</doi>
       <bibkey>wang-etal-2021-self</bibkey>
-      <pwccode url="https://github.com/junwang4/self-promotion-in-congress-tweets" additional="false">junwang4/self-promotion-in-congress-tweets</pwccode>
       <video href="2021.naacl-main.388.mp4"/>
+      <pwccode url="https://github.com/junwang4/self-promotion-in-congress-tweets" additional="false">junwang4/self-promotion-in-congress-tweets</pwccode>
     </paper>
     <paper id="389">
       <title>Profiling of Intertextuality in <fixed-case>L</fixed-case>atin Literature Using Word Embeddings</title>
@@ -5938,8 +5938,8 @@
       <url hash="6fa756fb">2021.naacl-main.389</url>
       <doi>10.18653/v1/2021.naacl-main.389</doi>
       <bibkey>burns-etal-2021-profiling</bibkey>
-      <pwccode url="https://github.com/quantitativecriticismlab/naacl-hlt-2021-latin-intertextuality" additional="false">quantitativecriticismlab/naacl-hlt-2021-latin-intertextuality</pwccode>
       <video href="2021.naacl-main.389.mp4"/>
+      <pwccode url="https://github.com/quantitativecriticismlab/naacl-hlt-2021-latin-intertextuality" additional="false">quantitativecriticismlab/naacl-hlt-2021-latin-intertextuality</pwccode>
     </paper>
     <paper id="390">
       <title>Identifying inherent disagreement in natural language inference</title>
@@ -5950,8 +5950,8 @@
       <url hash="7ddaf53b">2021.naacl-main.390</url>
       <doi>10.18653/v1/2021.naacl-main.390</doi>
       <bibkey>zhang-de-marneffe-2021-identifying</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
       <video href="2021.naacl-main.390.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/superglue">SuperGLUE</pwcdataset>
     </paper>
     <paper id="391">
       <title>Modeling Human Mental States with an Entity-based Narrative Graph</title>
@@ -5963,9 +5963,9 @@
       <url hash="e6036ba8">2021.naacl-main.391</url>
       <doi>10.18653/v1/2021.naacl-main.391</doi>
       <bibkey>lee-etal-2021-modeling</bibkey>
+      <video href="2021.naacl-main.391.mp4"/>
       <pwccode url="https://github.com/doug919/entity_based_narrative_graph" additional="false">doug919/entity_based_narrative_graph</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/desiredb">DesireDB</pwcdataset>
-      <video href="2021.naacl-main.391.mp4"/>
     </paper>
     <paper id="392">
       <title>A Simple and Efficient Multi-Task Learning Approach for Conditioned Dialogue Generation</title>
@@ -5989,12 +5989,12 @@
       <attachment type="OptionalSupplementaryData" hash="031c8096">2021.naacl-main.393.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.393</doi>
       <bibkey>krishna-etal-2021-hurdles</bibkey>
+      <video href="2021.naacl-main.393.mp4"/>
       <pwccode url="https://github.com/martiansideofthemoon/hurdles-longform-qa" additional="false">martiansideofthemoon/hurdles-longform-qa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/eli5">ELI5</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/kilt">KILT</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/pg-19">PG-19</pwcdataset>
-      <video href="2021.naacl-main.393.mp4"/>
     </paper>
     <paper id="394">
       <title><fixed-case>ENTRUST</fixed-case>: Argument Reframing with Language Models and Entailment</title>
@@ -6019,9 +6019,9 @@
       <url hash="18892798">2021.naacl-main.395</url>
       <doi>10.18653/v1/2021.naacl-main.395</doi>
       <bibkey>devaraj-etal-2021-paragraph</bibkey>
+      <video href="2021.naacl-main.395.mp4"/>
       <pwccode url="https://github.com/AshOlogn/Paragraph-level-Simplification-of-Medical-Texts" additional="false">AshOlogn/Paragraph-level-Simplification-of-Medical-Texts</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/newsela">Newsela</pwcdataset>
-      <video href="2021.naacl-main.395.mp4"/>
     </paper>
     <paper id="396">
       <title>An Empirical Study on Neural Keyphrase Generation</title>
@@ -6036,8 +6036,8 @@
       <url hash="29fd6db8">2021.naacl-main.396</url>
       <doi>10.18653/v1/2021.naacl-main.396</doi>
       <bibkey>meng-etal-2021-empirical</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
       <video href="2021.naacl-main.396.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/kp20k">KP20k</pwcdataset>
     </paper>
     <paper id="397">
       <title>Attention Head Masking for Inference Time Content Selection in Abstractive Summarization</title>
@@ -6062,10 +6062,10 @@
       <revision id="2" href="2021.naacl-main.398v2" hash="55e75ce4" date="2021-05-25">Updated a citation.</revision>
       <doi>10.18653/v1/2021.naacl-main.398</doi>
       <bibkey>zhong-etal-2021-factual</bibkey>
+      <video href="2021.naacl-main.398.mp4"/>
       <pwccode url="https://github.com/princeton-nlp/OptiPrompt" additional="true">princeton-nlp/OptiPrompt</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
-      <video href="2021.naacl-main.398.mp4"/>
     </paper>
     <paper id="399">
       <title>Evaluating Saliency Methods for Neural Language Models</title>
@@ -6076,10 +6076,11 @@
       <url hash="07dd3b47">2021.naacl-main.399</url>
       <doi>10.18653/v1/2021.naacl-main.399</doi>
       <bibkey>ding-koehn-2021-evaluating</bibkey>
+      <video href="2021.naacl-main.399.mp4"/>
       <pwccode url="https://github.com/shuoyangd/tarsius" additional="false">shuoyangd/tarsius</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/penn-treebank">Penn Treebank</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/winobias">WinoBias</pwcdataset>
-      <video href="2021.naacl-main.399.mp4"/>
     </paper>
     <paper id="400">
       <title>Contextualized Perturbation for Textual Adversarial Attack</title>
@@ -6095,11 +6096,11 @@
       <url hash="5aa24760">2021.naacl-main.400</url>
       <doi>10.18653/v1/2021.naacl-main.400</doi>
       <bibkey>li-etal-2021-contextualized</bibkey>
+      <video href="2021.naacl-main.400.mp4"/>
       <pwccode url="https://github.com/cookielee77/CLARE" additional="false">cookielee77/CLARE</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
-      <video href="2021.naacl-main.400.mp4"/>
     </paper>
     <paper id="401">
       <title><fixed-case>D</fixed-case>irect<fixed-case>P</fixed-case>robe: Studying Representations without Classifiers</title>
@@ -6110,8 +6111,8 @@
       <url hash="89156f56">2021.naacl-main.401</url>
       <doi>10.18653/v1/2021.naacl-main.401</doi>
       <bibkey>zhou-srikumar-2021-directprobe</bibkey>
-      <pwccode url="https://github.com/utahnlp/DirectProbe" additional="false">utahnlp/DirectProbe</pwccode>
       <video href="2021.naacl-main.401.mp4"/>
+      <pwccode url="https://github.com/utahnlp/DirectProbe" additional="false">utahnlp/DirectProbe</pwccode>
     </paper>
     <paper id="402">
       <title>Evaluating the Values of Sources in Transfer Learning</title>
@@ -6122,12 +6123,12 @@
       <url hash="8bf9997f">2021.naacl-main.402</url>
       <doi>10.18653/v1/2021.naacl-main.402</doi>
       <bibkey>parvez-chang-2021-evaluating</bibkey>
+      <video href="2021.naacl-main.402.mp4"/>
       <pwccode url="https://github.com/rizwan09/NLPDV" additional="false">rizwan09/NLPDV</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/qnli">QNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/universal-dependencies">Universal Dependencies</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/xnli">XNLI</pwcdataset>
-      <video href="2021.naacl-main.402.mp4"/>
     </paper>
     <paper id="403">
       <title>Too Much in Common: Shifting of Embeddings in Transformer Language Models and its Implications</title>
@@ -6139,8 +6140,8 @@
       <url hash="c5528f55">2021.naacl-main.403</url>
       <doi>10.18653/v1/2021.naacl-main.403</doi>
       <bibkey>bis-etal-2021-much</bibkey>
-      <pwccode url="https://github.com/danielbis/toomuchincommon" additional="false">danielbis/toomuchincommon</pwccode>
       <video href="2021.naacl-main.403.mp4"/>
+      <pwccode url="https://github.com/danielbis/toomuchincommon" additional="false">danielbis/toomuchincommon</pwccode>
     </paper>
     <paper id="404">
       <title>On the Inductive Bias of Masked Language Modeling: From Statistical to Syntactic Dependencies</title>
@@ -6151,9 +6152,9 @@
       <url hash="e38eece1">2021.naacl-main.404</url>
       <doi>10.18653/v1/2021.naacl-main.404</doi>
       <bibkey>zhang-hashimoto-2021-inductive</bibkey>
+      <video href="2021.naacl-main.404.mp4"/>
       <pwccode url="https://github.com/tatsu-lab/mlm_inductive_bias" additional="false">tatsu-lab/mlm_inductive_bias</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.404.mp4"/>
     </paper>
     <paper id="405">
       <title>Limitations of Autoregressive Models and Their Alternatives</title>
@@ -6184,8 +6185,8 @@
       <url hash="4e971e21">2021.naacl-main.406</url>
       <doi>10.18653/v1/2021.naacl-main.406</doi>
       <bibkey>gu-etal-2021-transformer</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <video href="2021.naacl-main.406.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
     </paper>
     <paper id="407">
       <title>Revisiting Simple Neural Probabilistic Language Models</title>
@@ -6196,10 +6197,11 @@
       <url hash="6e0306af">2021.naacl-main.407</url>
       <doi>10.18653/v1/2021.naacl-main.407</doi>
       <bibkey>sun-iyyer-2021-revisiting</bibkey>
+      <video href="2021.naacl-main.407.mp4"/>
       <pwccode url="https://github.com/SimengSun/revisit-nplm" additional="false">SimengSun/revisit-nplm</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/lambada">LAMBADA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikitext-103">WikiText-103</pwcdataset>
-      <video href="2021.naacl-main.407.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wikitext-2">WikiText-2</pwcdataset>
     </paper>
     <paper id="408">
       <title><fixed-case>R</fixed-case>ead<fixed-case>T</fixed-case>wice: Reading Very Large Documents with Memories</title>
@@ -6214,10 +6216,10 @@
       <url hash="7995febf">2021.naacl-main.408</url>
       <doi>10.18653/v1/2021.naacl-main.408</doi>
       <bibkey>zemlyanskiy-etal-2021-readtwice</bibkey>
+      <video href="2021.naacl-main.408.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/narrativeqa">NarrativeQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.naacl-main.408.mp4"/>
     </paper>
     <paper id="409">
       <title><fixed-case>SCRIPT</fixed-case>: Self-Critic <fixed-case>P</fixed-case>re<fixed-case>T</fixed-case>raining of Transformers</title>
@@ -6230,11 +6232,11 @@
       <url hash="54cf3f04">2021.naacl-main.409</url>
       <doi>10.18653/v1/2021.naacl-main.409</doi>
       <bibkey>nijkamp-etal-2021-script</bibkey>
+      <video href="2021.naacl-main.409.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/openwebtext">OpenWebText</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webtext">WebText</pwcdataset>
-      <video href="2021.naacl-main.409.mp4"/>
     </paper>
     <paper id="410">
       <title>Learning How to Ask: Querying <fixed-case>LM</fixed-case>s with Mixtures of Soft Prompts</title>
@@ -6248,11 +6250,11 @@
       <award>Best Short Paper</award>
       <doi>10.18653/v1/2021.naacl-main.410</doi>
       <bibkey>qin-eisner-2021-learning</bibkey>
+      <video href="2021.naacl-main.410.mp4"/>
       <pwccode url="https://github.com/hiaoxui/soft-prompts" additional="true">hiaoxui/soft-prompts</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/lama">LAMA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/t-rex">T-REx</pwcdataset>
-      <video href="2021.naacl-main.410.mp4"/>
     </paper>
     <paper id="411">
       <title>Nutri-bullets Hybrid: Consensual Multi-document Summarization</title>
@@ -6267,8 +6269,8 @@
       <attachment type="OptionalSupplementaryData" hash="94229b49">2021.naacl-main.411.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.411</doi>
       <bibkey>shah-etal-2021-nutri</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/healthline">Healthline</pwcdataset>
       <video href="2021.naacl-main.411.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/healthline">Healthline</pwcdataset>
     </paper>
     <paper id="412">
       <title><fixed-case>AVA</fixed-case>: an Automatic e<fixed-case>V</fixed-case>aluation Approach for Question Answering Systems</title>
@@ -6295,8 +6297,8 @@
       <attachment type="OptionalSupplementaryCode" hash="65e3c865">2021.naacl-main.413.OptionalSupplementaryCode.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.413</doi>
       <bibkey>subramanian-etal-2021-spanpredict</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <video href="2021.naacl-main.413.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
     </paper>
     <paper id="414">
       <title>Text Editing by Command</title>
@@ -6312,9 +6314,9 @@
       <url hash="5ea5f670">2021.naacl-main.414</url>
       <doi>10.18653/v1/2021.naacl-main.414</doi>
       <bibkey>faltings-etal-2021-text</bibkey>
+      <video href="2021.naacl-main.414.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/wikidocedits">WikiDocEdits</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikiatomicedits">WikiAtomicEdits</pwcdataset>
-      <video href="2021.naacl-main.414.mp4"/>
     </paper>
     <paper id="415">
       <title>A Deep Metric Learning Approach to Account Linking</title>
@@ -6328,8 +6330,8 @@
       <url hash="cce73231">2021.naacl-main.415</url>
       <doi>10.18653/v1/2021.naacl-main.415</doi>
       <bibkey>khan-etal-2021-deep</bibkey>
-      <pwccode url="https://github.com/noa/naacl2021" additional="true">noa/naacl2021</pwccode>
       <video href="2021.naacl-main.415.mp4"/>
+      <pwccode url="https://github.com/noa/naacl2021" additional="true">noa/naacl2021</pwccode>
     </paper>
     <paper id="416">
       <title>Improving Factual Completeness and Consistency of Image-to-Text Radiology Report Generation</title>
@@ -6343,8 +6345,8 @@
       <url hash="e4e72a3e">2021.naacl-main.416</url>
       <doi>10.18653/v1/2021.naacl-main.416</doi>
       <bibkey>miura-etal-2021-improving</bibkey>
-      <pwccode url="https://github.com/ysmiura/ifcc" additional="true">ysmiura/ifcc</pwccode>
       <video href="2021.naacl-main.416.mp4"/>
+      <pwccode url="https://github.com/ysmiura/ifcc" additional="true">ysmiura/ifcc</pwccode>
     </paper>
     <paper id="417">
       <title>Multimodal End-to-End Sparse Model for Emotion Recognition</title>
@@ -6357,10 +6359,10 @@
       <url hash="1166f180">2021.naacl-main.417</url>
       <doi>10.18653/v1/2021.naacl-main.417</doi>
       <bibkey>dai-etal-2021-multimodal</bibkey>
+      <video href="2021.naacl-main.417.mp4"/>
       <pwccode url="https://github.com/wenliangdai/Multimodal-End2end-Sparse" additional="false">wenliangdai/Multimodal-End2end-Sparse</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cmu-mosei">CMU-MOSEI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
-      <video href="2021.naacl-main.417.mp4"/>
     </paper>
     <paper id="418">
       <title><fixed-case>MIMOQA</fixed-case>: Multimodal Input Multimodal Output Question Answering</title>
@@ -6375,6 +6377,7 @@
       <url hash="9da2260b">2021.naacl-main.418</url>
       <doi>10.18653/v1/2021.naacl-main.418</doi>
       <bibkey>singh-etal-2021-mimoqa</bibkey>
+      <video href="2021.naacl-main.418.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/manymodalqa">ManyModalQA</pwcdataset>
@@ -6382,7 +6385,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/tvqa">TVQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.naacl-main.418.mp4"/>
     </paper>
     <paper id="419">
       <title><fixed-case>OCID</fixed-case>-Ref: A 3<fixed-case>D</fixed-case> Robotic Dataset With Embodied Language For Clutter Scene Grounding</title>
@@ -6399,11 +6401,11 @@
       <attachment type="OptionalSupplementaryData" hash="be7e740e">2021.naacl-main.419.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.419</doi>
       <bibkey>wang-etal-2021-ocid</bibkey>
+      <video href="2021.naacl-main.419.mp4"/>
       <pwccode url="https://github.com/lluma/OCID-Ref" additional="false">lluma/OCID-Ref</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cops-ref">Cops-Ref</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ocid">OCID</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/refcoco">RefCOCO</pwcdataset>
-      <video href="2021.naacl-main.419.mp4"/>
     </paper>
     <paper id="420">
       <title>Unsupervised Vision-and-Language Pre-training Without Parallel Images and Captions</title>
@@ -6419,11 +6421,11 @@
       <attachment type="OptionalSupplementaryData" hash="1fa89178">2021.naacl-main.420.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.420</doi>
       <bibkey>li-etal-2021-unsupervised</bibkey>
+      <video href="2021.naacl-main.420.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/refcoco">RefCOCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering-v2-0">Visual Question Answering v2.0</pwcdataset>
-      <video href="2021.naacl-main.420.mp4"/>
     </paper>
     <paper id="421">
       <title>Multitasking Inhibits Semantic Drift</title>
@@ -6448,10 +6450,10 @@
       <url hash="81ce6752">2021.naacl-main.422</url>
       <doi>10.18653/v1/2021.naacl-main.422</doi>
       <bibkey>ilharco-etal-2021-probing</bibkey>
+      <video href="2021.naacl-main.422.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-genome">Visual Genome</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/visual-question-answering">Visual Question Answering</pwcdataset>
-      <video href="2021.naacl-main.422.mp4"/>
     </paper>
     <paper id="423">
       <title><fixed-case>BBAEG</fixed-case>: Towards <fixed-case>BERT</fixed-case>-based Biomedical Adversarial Example Generation for Text Classification</title>
@@ -6462,8 +6464,8 @@
       <attachment type="OptionalSupplementaryData" hash="242b095d">2021.naacl-main.423.OptionalSupplementaryData.pdf</attachment>
       <doi>10.18653/v1/2021.naacl-main.423</doi>
       <bibkey>mondal-2021-bbaeg</bibkey>
-      <pwccode url="https://github.com/Ishani-Mondal/BBAEG" additional="false">Ishani-Mondal/BBAEG</pwccode>
       <video href="2021.naacl-main.423.mp4"/>
+      <pwccode url="https://github.com/Ishani-Mondal/BBAEG" additional="false">Ishani-Mondal/BBAEG</pwccode>
     </paper>
     <paper id="424">
       <title>Targeted Adversarial Training for Natural Language Understanding</title>
@@ -6478,11 +6480,11 @@
       <url hash="25498895">2021.naacl-main.424</url>
       <doi>10.18653/v1/2021.naacl-main.424</doi>
       <bibkey>pereira-etal-2021-targeted</bibkey>
+      <video href="2021.naacl-main.424.mp4"/>
       <pwccode url="https://github.com/namisan/mt-dnn" additional="false">namisan/mt-dnn</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mrpc">MRPC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
-      <video href="2021.naacl-main.424.mp4"/>
     </paper>
     <paper id="425">
       <title>Latent-Optimized Adversarial Neural Transfer for Sarcasm Detection</title>
@@ -6495,9 +6497,9 @@
       <url hash="b4a7bf64">2021.naacl-main.425</url>
       <doi>10.18653/v1/2021.naacl-main.425</doi>
       <bibkey>guo-etal-2021-latent</bibkey>
+      <video href="2021.naacl-main.425.mp4"/>
       <pwccode url="https://github.com/guoxuxu/LOANT" additional="false">guoxuxu/LOANT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/isarcasm">iSarcasm</pwcdataset>
-      <video href="2021.naacl-main.425.mp4"/>
     </paper>
     <paper id="426">
       <title>Self-training Improves Pre-training for Natural Language Understanding</title>
@@ -6533,9 +6535,9 @@
       <url hash="86a3ca4d">2021.naacl-main.427</url>
       <doi>10.18653/v1/2021.naacl-main.427</doi>
       <bibkey>zhang-etal-2021-supporting</bibkey>
+      <video href="2021.naacl-main.427.mp4"/>
       <pwccode url="https://github.com/amazon-research/sccl" additional="true">amazon-research/sccl</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
-      <video href="2021.naacl-main.427.mp4"/>
     </paper>
     <paper id="428">
       <title><fixed-case>TITA</fixed-case>: A Two-stage Interaction and Topic-Aware Text Matching Model</title>
@@ -6564,12 +6566,12 @@
       <url hash="4b3cb79b">2021.naacl-main.429</url>
       <doi>10.18653/v1/2021.naacl-main.429</doi>
       <bibkey>liu-etal-2021-neural</bibkey>
+      <video href="2021.naacl-main.429.mp4"/>
       <pwccode url="https://github.com/thunlp/VERNet" additional="false">thunlp/VERNet</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2014-shared-task-grammatical-error">CoNLL-2014 Shared Task: Grammatical Error Correction</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/fce">FCE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/jfleg">JFLEG</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/locness-corpus">WI-LOCNESS</pwcdataset>
-      <video href="2021.naacl-main.429.mp4"/>
     </paper>
     <paper id="430">
       <title>Neural Network Surgery: Injecting Data Patterns into Pre-trained Models with Minimal Instance-wise Side Effects</title>
@@ -6583,10 +6585,10 @@
       <url hash="40c9a728">2021.naacl-main.430</url>
       <doi>10.18653/v1/2021.naacl-main.430</doi>
       <bibkey>zhang-etal-2021-neural</bibkey>
+      <video href="2021.naacl-main.430.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-main.430.mp4"/>
     </paper>
     <paper id="431">
       <title>Discrete Argument Representation Learning for Interactive Argument Pair Identification</title>
@@ -6616,8 +6618,8 @@
       <url hash="41a63720">2021.naacl-main.432</url>
       <doi>10.18653/v1/2021.naacl-main.432</doi>
       <bibkey>lee-etal-2021-unifying</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/basil">BASIL</pwcdataset>
       <video href="2021.naacl-main.432.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/basil">BASIL</pwcdataset>
     </paper>
     <paper id="433">
       <title>Frustratingly Easy Edit-based Linguistic Steganography with a Masked Language Model</title>
@@ -6629,8 +6631,8 @@
       <url hash="bd03723f">2021.naacl-main.433</url>
       <doi>10.18653/v1/2021.naacl-main.433</doi>
       <bibkey>ueoka-etal-2021-frustratingly</bibkey>
-      <pwccode url="https://github.com/ku-nlp/steganography-with-masked-lm" additional="false">ku-nlp/steganography-with-masked-lm</pwccode>
       <video href="2021.naacl-main.433.mp4"/>
+      <pwccode url="https://github.com/ku-nlp/steganography-with-masked-lm" additional="false">ku-nlp/steganography-with-masked-lm</pwccode>
     </paper>
     <paper id="434">
       <title>Few-Shot Text Classification with Triplet Networks, Data Augmentation, and Curriculum Learning</title>
@@ -6644,9 +6646,9 @@
       <url hash="af60bf89">2021.naacl-main.434</url>
       <doi>10.18653/v1/2021.naacl-main.434</doi>
       <bibkey>wei-etal-2021-shot</bibkey>
+      <video href="2021.naacl-main.434.mp4"/>
       <pwccode url="https://github.com/jasonwei20/triplet-loss" additional="false">jasonwei20/triplet-loss</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
-      <video href="2021.naacl-main.434.mp4"/>
     </paper>
     <paper id="435">
       <title>Do <fixed-case>RNN</fixed-case> States Encode Abstract Phonological Alternations?</title>
@@ -6703,8 +6705,8 @@
       <url hash="f1141938">2021.naacl-main.438</url>
       <doi>10.18653/v1/2021.naacl-main.438</doi>
       <bibkey>higashiyama-etal-2021-user</bibkey>
-      <pwccode url="https://github.com/shigashiyama/jlexnorm" additional="false">shigashiyama/jlexnorm</pwccode>
       <video href="2021.naacl-main.438.mp4"/>
+      <pwccode url="https://github.com/shigashiyama/jlexnorm" additional="false">shigashiyama/jlexnorm</pwccode>
     </paper>
     <paper id="439">
       <title><fixed-case>GPT</fixed-case> Perdetry Test: Generating new meanings for new words</title>
@@ -6733,8 +6735,8 @@
       <attachment type="OptionalSupplementaryData" hash="7bb47dd3">2021.naacl-main.440.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.440</doi>
       <bibkey>li-etal-2021-universal</bibkey>
-      <pwccode url="https://github.com/pkucoli/ust" additional="false">pkucoli/ust</pwccode>
       <video href="2021.naacl-main.440.mp4"/>
+      <pwccode url="https://github.com/pkucoli/ust" additional="false">pkucoli/ust</pwccode>
     </paper>
     <paper id="441">
       <title><fixed-case>S</fixed-case>hadow<fixed-case>GNN</fixed-case>: Graph Projection Neural Network for Text-to-<fixed-case>SQL</fixed-case> Parser</title>
@@ -6751,9 +6753,9 @@
       <url hash="d3fef69d">2021.naacl-main.441</url>
       <doi>10.18653/v1/2021.naacl-main.441</doi>
       <bibkey>chen-etal-2021-shadowgnn</bibkey>
+      <video href="2021.naacl-main.441.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/scan">SCAN</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/spider-1">SPIDER</pwcdataset>
-      <video href="2021.naacl-main.441.mp4"/>
     </paper>
     <paper id="442">
       <title>Contextualized and Generalized Sentence Representations by Contrastive Self-Supervised Learning: A Case Study on Discourse Relation Analysis</title>
@@ -6764,8 +6766,8 @@
       <url hash="76a6090e">2021.naacl-main.442</url>
       <doi>10.18653/v1/2021.naacl-main.442</doi>
       <bibkey>kiyomaru-kurohashi-2021-contextualized</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
       <video href="2021.naacl-main.442.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/bookcorpus">BookCorpus</pwcdataset>
     </paper>
     <paper id="443">
       <title><fixed-case>AMR</fixed-case> Parsing with Action-Pointer Transformer</title>
@@ -6778,9 +6780,9 @@
       <url hash="6f18c4ee">2021.naacl-main.443</url>
       <doi>10.18653/v1/2021.naacl-main.443</doi>
       <bibkey>zhou-etal-2021-amr</bibkey>
+      <video href="2021.naacl-main.443.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/ldc2017t10">LDC2017T10</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ldc2020t02">LDC2020T02</pwcdataset>
-      <video href="2021.naacl-main.443.mp4"/>
     </paper>
     <paper id="444">
       <title><fixed-case>NL</fixed-case>-<fixed-case>EDIT</fixed-case>: Correcting Semantic Parse Errors through Natural Language Interaction</title>
@@ -6795,8 +6797,8 @@
       <url hash="623a4ff3">2021.naacl-main.444</url>
       <doi>10.18653/v1/2021.naacl-main.444</doi>
       <bibkey>elgohary-etal-2021-nl</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/splash">SPLASH</pwcdataset>
       <video href="2021.naacl-main.444.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/splash">SPLASH</pwcdataset>
     </paper>
     <paper id="445">
       <title>Unsupervised Concept Representation Learning for Length-Varying Text Similarity</title>
@@ -6826,8 +6828,8 @@
       <url hash="78632b55">2021.naacl-main.446</url>
       <doi>10.18653/v1/2021.naacl-main.446</doi>
       <bibkey>zhan-etal-2021-augmenting</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
       <video href="2021.naacl-main.446.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
     </paper>
     <paper id="447">
       <title>Adversarial Self-Supervised Learning for Out-of-Domain Detection</title>
@@ -6841,9 +6843,9 @@
       <url hash="a04ad55d">2021.naacl-main.447</url>
       <doi>10.18653/v1/2021.naacl-main.447</doi>
       <bibkey>zeng-etal-2021-adversarial</bibkey>
+      <video href="2021.naacl-main.447.mp4"/>
       <pwccode url="https://github.com/parzival27/adversarial-self-supervised-out-of-domain-detection" additional="false">parzival27/adversarial-self-supervised-out-of-domain-detection</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clinc150">CLINC150</pwcdataset>
-      <video href="2021.naacl-main.447.mp4"/>
     </paper>
     <paper id="448">
       <title>Leveraging Slot Descriptions for Zero-Shot Cross-Domain Dialogue <fixed-case>S</fixed-case>tate<fixed-case>T</fixed-case>racking</title>
@@ -6862,8 +6864,8 @@
       <url hash="ce6a4474">2021.naacl-main.448</url>
       <doi>10.18653/v1/2021.naacl-main.448</doi>
       <bibkey>lin-etal-2021-leveraging</bibkey>
-      <pwccode url="https://github.com/facebookresearch/Zero-Shot-DST" additional="false">facebookresearch/Zero-Shot-DST</pwccode>
       <video href="2021.naacl-main.448.mp4"/>
+      <pwccode url="https://github.com/facebookresearch/Zero-Shot-DST" additional="false">facebookresearch/Zero-Shot-DST</pwccode>
     </paper>
     <paper id="449">
       <title>Hierarchical Transformer for Task Oriented Dialog Systems</title>
@@ -6875,9 +6877,9 @@
       <url hash="c615c674">2021.naacl-main.449</url>
       <doi>10.18653/v1/2021.naacl-main.449</doi>
       <bibkey>santra-etal-2021-hierarchical</bibkey>
+      <video href="2021.naacl-main.449.mp4"/>
       <pwccode url="https://github.com/bsantraigi/hier-transformer-pytorch" additional="true">bsantraigi/hier-transformer-pytorch</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
-      <video href="2021.naacl-main.449.mp4"/>
     </paper>
     <paper id="450">
       <title>Measuring the ‘<fixed-case>I</fixed-case> don’t know’ Problem through the Lens of <fixed-case>G</fixed-case>ricean Quantity</title>
@@ -6888,9 +6890,9 @@
       <url hash="d0d2e04d">2021.naacl-main.450</url>
       <doi>10.18653/v1/2021.naacl-main.450</doi>
       <bibkey>khayrallah-sedoc-2021-measuring</bibkey>
+      <video href="2021.naacl-main.450.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flores">FLoRes</pwcdataset>
-      <video href="2021.naacl-main.450.mp4"/>
     </paper>
     <paper id="451">
       <title><fixed-case>RTFE</fixed-case>: A Recursive Temporal Fact Embedding Framework for Temporal Knowledge Graph Completion</title>
@@ -6906,8 +6908,8 @@
       <url hash="232743f8">2021.naacl-main.451</url>
       <doi>10.18653/v1/2021.naacl-main.451</doi>
       <bibkey>xu-etal-2021-rtfe</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
       <video href="2021.naacl-main.451.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/icews">ICEWS</pwcdataset>
     </paper>
     <paper id="452">
       <title>Open Hierarchical Relation Extraction</title>
@@ -6924,9 +6926,9 @@
       <url hash="7ce61d5c">2021.naacl-main.452</url>
       <doi>10.18653/v1/2021.naacl-main.452</doi>
       <bibkey>zhang-etal-2021-open</bibkey>
+      <video href="2021.naacl-main.452.mp4"/>
       <pwccode url="https://github.com/thunlp/ohre" additional="false">thunlp/ohre</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fewrel">FewRel</pwcdataset>
-      <video href="2021.naacl-main.452.mp4"/>
     </paper>
     <paper id="453">
       <title>Jointly Extracting Explicit and Implicit Relational Triples with Reasoning Pattern Enhanced Binary Pointer Network</title>
@@ -6955,8 +6957,8 @@
       <url hash="80a493cf">2021.naacl-main.454</url>
       <doi>10.18653/v1/2021.naacl-main.454</doi>
       <bibkey>zhou-etal-2021-multi</bibkey>
-      <pwccode url="https://github.com/11zhouxuan/multi_grained_kd_ner" additional="false">11zhouxuan/multi_grained_kd_ner</pwccode>
       <video href="2021.naacl-main.454.mp4"/>
+      <pwccode url="https://github.com/11zhouxuan/multi_grained_kd_ner" additional="false">11zhouxuan/multi_grained_kd_ner</pwccode>
     </paper>
     <paper id="455">
       <title><fixed-case>SGG</fixed-case>: Learning to Select, Guide, and Generate for Keyphrase Generation</title>
@@ -6971,8 +6973,8 @@
       <url hash="853573e7">2021.naacl-main.455</url>
       <doi>10.18653/v1/2021.naacl-main.455</doi>
       <bibkey>zhao-etal-2021-sgg</bibkey>
-      <pwccode url="https://github.com/jd-ai-research-nlp/sgg" additional="false">jd-ai-research-nlp/sgg</pwccode>
       <video href="2021.naacl-main.455.mp4"/>
+      <pwccode url="https://github.com/jd-ai-research-nlp/sgg" additional="false">jd-ai-research-nlp/sgg</pwccode>
     </paper>
     <paper id="456">
       <title>Towards Sentiment and Emotion aided Multi-modal Speech Act Classification in <fixed-case>T</fixed-case>witter</title>
@@ -6997,9 +6999,9 @@
       <url hash="2a5a5564">2021.naacl-main.457</url>
       <doi>10.18653/v1/2021.naacl-main.457</doi>
       <bibkey>long-etal-2021-generative</bibkey>
+      <video href="2021.naacl-main.457.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
-      <video href="2021.naacl-main.457.mp4"/>
     </paper>
     <paper id="458">
       <title>Non-Autoregressive Translation by Learning Target Categorical Codes</title>
@@ -7014,9 +7016,9 @@
       <url hash="f1787812">2021.naacl-main.458</url>
       <doi>10.18653/v1/2021.naacl-main.458</doi>
       <bibkey>bao-etal-2021-non</bibkey>
+      <video href="2021.naacl-main.458.mp4"/>
       <pwccode url="https://github.com/baoy-nlp/CNAT" additional="false">baoy-nlp/CNAT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
-      <video href="2021.naacl-main.458.mp4"/>
     </paper>
     <paper id="459">
       <title>Training Data Augmentation for Code-Mixed Translation</title>
@@ -7028,9 +7030,9 @@
       <url hash="f7ca05bd">2021.naacl-main.459</url>
       <doi>10.18653/v1/2021.naacl-main.459</doi>
       <bibkey>gupta-etal-2021-training</bibkey>
+      <video href="2021.naacl-main.459.mp4"/>
       <pwccode url="https://github.com/shruikan20/spoken-tutorial-dataset" additional="false">shruikan20/spoken-tutorial-dataset</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/phinc">PHINC</pwcdataset>
-      <video href="2021.naacl-main.459.mp4"/>
     </paper>
     <paper id="460">
       <title>Rethinking Perturbations in Encoder-Decoders for Fast Training</title>
@@ -7041,10 +7043,10 @@
       <url hash="daef1039">2021.naacl-main.460</url>
       <doi>10.18653/v1/2021.naacl-main.460</doi>
       <bibkey>takase-kiyono-2021-rethinking</bibkey>
+      <video href="2021.naacl-main.460.mp4"/>
       <pwccode url="https://github.com/takase/rethink_perturbations" additional="false">takase/rethink_perturbations</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/duc-2004">DUC 2004</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wmt-2014">WMT 2014</pwcdataset>
-      <video href="2021.naacl-main.460.mp4"/>
     </paper>
     <paper id="461">
       <title>Context-aware Decoder for Neural Machine Translation using a Target-side Document-Level Language Model</title>
@@ -7069,8 +7071,8 @@
       <url hash="065edd72">2021.naacl-main.462</url>
       <doi>10.18653/v1/2021.naacl-main.462</doi>
       <bibkey>nguyen-son-etal-2021-machine</bibkey>
-      <pwccode url="https://github.com/quocnsh/machine_translation_detection" additional="false">quocnsh/machine_translation_detection</pwccode>
       <video href="2021.naacl-main.462.mp4"/>
+      <pwccode url="https://github.com/quocnsh/machine_translation_detection" additional="false">quocnsh/machine_translation_detection</pwccode>
     </paper>
     <paper id="463">
       <title><fixed-case>TR</fixed-case>-<fixed-case>BERT</fixed-case>: Dynamic Token Reduction for Accelerating <fixed-case>BERT</fixed-case> Inference</title>
@@ -7083,6 +7085,7 @@
       <url hash="e279e6ef">2021.naacl-main.463</url>
       <doi>10.18653/v1/2021.naacl-main.463</doi>
       <bibkey>ye-etal-2021-tr</bibkey>
+      <video href="2021.naacl-main.463.mp4"/>
       <pwccode url="https://github.com/thunlp/TR-BERT" additional="false">thunlp/TR-BERT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
@@ -7090,7 +7093,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/race">RACE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihop">WikiHop</pwcdataset>
-      <video href="2021.naacl-main.463.mp4"/>
     </paper>
     <paper id="464">
       <title>Breadth First Reasoning Graph for Multi-hop Question Answering</title>
@@ -7101,9 +7103,9 @@
       <url hash="3dc83a3f">2021.naacl-main.464</url>
       <doi>10.18653/v1/2021.naacl-main.464</doi>
       <bibkey>huang-yang-2021-breadth</bibkey>
+      <video href="2021.naacl-main.464.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikihop">WikiHop</pwcdataset>
-      <video href="2021.naacl-main.464.mp4"/>
     </paper>
     <paper id="465">
       <title>Improving Zero-Shot Cross-lingual Transfer for Multilingual Question Answering over Knowledge Graph</title>
@@ -7117,8 +7119,8 @@
       <url hash="3d82cf2e">2021.naacl-main.465</url>
       <doi>10.18653/v1/2021.naacl-main.465</doi>
       <bibkey>zhou-etal-2021-improving</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/dbpedia">DBpedia</pwcdataset>
       <video href="2021.naacl-main.465.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/dbpedia">DBpedia</pwcdataset>
     </paper>
     <paper id="466">
       <title><fixed-case>R</fixed-case>ocket<fixed-case>QA</fixed-case>: An Optimized Training Approach to Dense Passage Retrieval for Open-Domain Question Answering</title>
@@ -7136,11 +7138,11 @@
       <url hash="4f7c8b53">2021.naacl-main.466</url>
       <doi>10.18653/v1/2021.naacl-main.466</doi>
       <bibkey>qu-etal-2021-rocketqa</bibkey>
+      <video href="2021.naacl-main.466.mp4"/>
       <pwccode url="https://github.com/paddlepaddle/rocketqa" additional="false">paddlepaddle/rocketqa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mrqa-2019">MRQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
-      <video href="2021.naacl-main.466.mp4"/>
     </paper>
     <paper id="467">
       <title><fixed-case>DAGN</fixed-case>: Discourse-Aware Graph Network for Logical Reasoning</title>
@@ -7154,12 +7156,12 @@
       <url hash="32c3fcb0">2021.naacl-main.467</url>
       <doi>10.18653/v1/2021.naacl-main.467</doi>
       <bibkey>huang-etal-2021-dagn</bibkey>
+      <video href="2021.naacl-main.467.mp4"/>
       <pwccode url="https://github.com/Eleanor-H/DAGN" additional="false">Eleanor-H/DAGN</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/logiqa">LogiQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/reclor">ReClor</pwcdataset>
-      <video href="2021.naacl-main.467.mp4"/>
     </paper>
     <paper id="468">
       <title>Designing a Minimal Retrieve-and-Read System for Open-Domain Question Answering</title>
@@ -7170,10 +7172,10 @@
       <url hash="cfca1071">2021.naacl-main.468</url>
       <doi>10.18653/v1/2021.naacl-main.468</doi>
       <bibkey>yang-seo-2021-designing</bibkey>
+      <video href="2021.naacl-main.468.mp4"/>
       <pwccode url="https://github.com/clovaai/minimal-rnr-qa" additional="false">clovaai/minimal-rnr-qa</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/natural-questions">Natural Questions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.naacl-main.468.mp4"/>
     </paper>
     <paper id="469">
       <title>Unsupervised Multi-hop Question Answering by Question Generation</title>
@@ -7187,10 +7189,10 @@
       <url hash="d59cda74">2021.naacl-main.469</url>
       <doi>10.18653/v1/2021.naacl-main.469</doi>
       <bibkey>pan-etal-2021-unsupervised</bibkey>
+      <video href="2021.naacl-main.469.mp4"/>
       <pwccode url="https://github.com/teacherpeterpan/Unsupervised-Multi-hop-QA" additional="false">teacherpeterpan/Unsupervised-Multi-hop-QA</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/hotpotqa">HotpotQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hybridqa">HybridQA</pwcdataset>
-      <video href="2021.naacl-main.469.mp4"/>
     </paper>
     <paper id="470">
       <title>Sliding Selector Network with Dynamic Memory for Extractive Summarization of Long Documents</title>
@@ -7201,8 +7203,8 @@
       <url hash="9077b950">2021.naacl-main.470</url>
       <doi>10.18653/v1/2021.naacl-main.470</doi>
       <bibkey>cui-hu-2021-sliding</bibkey>
-      <pwccode url="https://github.com/pcui-nlp/ssn_dm" additional="false">pcui-nlp/ssn_dm</pwccode>
       <video href="2021.naacl-main.470.mp4"/>
+      <pwccode url="https://github.com/pcui-nlp/ssn_dm" additional="false">pcui-nlp/ssn_dm</pwccode>
     </paper>
     <paper id="471">
       <title><fixed-case>A</fixed-case>dapt<fixed-case>S</fixed-case>um: Towards Low-Resource Domain Adaptation for Abstractive Summarization</title>
@@ -7215,11 +7217,11 @@
       <attachment type="OptionalSupplementaryCode" hash="45a145ff">2021.naacl-main.471.OptionalSupplementaryCode.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.471</doi>
       <bibkey>yu-etal-2021-adaptsum</bibkey>
+      <video href="2021.naacl-main.471.mp4"/>
       <pwccode url="https://github.com/TysonYu/AdaptSum" additional="false">TysonYu/AdaptSum</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/reddit-tifu">Reddit TIFU</pwcdataset>
-      <video href="2021.naacl-main.471.mp4"/>
     </paper>
     <paper id="472">
       <title><fixed-case>QMS</fixed-case>um: A New Benchmark for Query-based Multi-domain Meeting Summarization</title>
@@ -7240,9 +7242,9 @@
       <attachment type="OptionalSupplementaryData" hash="b6b58dbe">2021.naacl-main.472.OptionalSupplementaryData.zip</attachment>
       <doi>10.18653/v1/2021.naacl-main.472</doi>
       <bibkey>zhong-etal-2021-qmsum</bibkey>
+      <video href="2021.naacl-main.472.mp4"/>
       <pwccode url="https://github.com/Yale-LILY/QMSum" additional="false">Yale-LILY/QMSum</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/qmsum">QMSum</pwcdataset>
-      <video href="2021.naacl-main.472.mp4"/>
     </paper>
     <paper id="473">
       <title><fixed-case>MM</fixed-case>-<fixed-case>AVS</fixed-case>: A Full-Scale Dataset for Multi-modal Summarization</title>
@@ -7254,9 +7256,9 @@
       <url hash="2e840649">2021.naacl-main.473</url>
       <doi>10.18653/v1/2021.naacl-main.473</doi>
       <bibkey>fu-etal-2021-mm</bibkey>
+      <video href="2021.naacl-main.473.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/how2">How2</pwcdataset>
-      <video href="2021.naacl-main.473.mp4"/>
     </paper>
     <paper id="474">
       <title><fixed-case>M</fixed-case>edia<fixed-case>S</fixed-case>um: A Large-scale Media Interview Dataset for Dialogue Summarization</title>
@@ -7270,12 +7272,12 @@
       <attachment type="OptionalSupplementaryMaterial" hash="a0ef8c06">2021.naacl-main.474.OptionalSupplementaryMaterial.pdf</attachment>
       <doi>10.18653/v1/2021.naacl-main.474</doi>
       <bibkey>zhu-etal-2021-mediasum</bibkey>
+      <video href="2021.naacl-main.474.mp4"/>
       <pwccode url="https://github.com/zcgzcgzcg1/MediaSum" additional="false">zcgzcgzcg1/MediaSum</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/crd3">CRD3</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/interview">Interview</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multiwoz">MultiWOZ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/samsum-corpus">SAMSum Corpus</pwcdataset>
-      <video href="2021.naacl-main.474.mp4"/>
     </paper>
     <paper id="475">
       <title>Improving Faithfulness in Abstractive Summarization with Contrast Candidate Generation and Selection</title>
@@ -7340,8 +7342,8 @@
       <url hash="478af461">2021.naacl-demos.1</url>
       <doi>10.18653/v1/2021.naacl-demos.1</doi>
       <bibkey>nguyen-nguyen-2021-phonlp</bibkey>
-      <pwccode url="https://github.com/VinAIResearch/PhoNLP" additional="false">VinAIResearch/PhoNLP</pwccode>
       <video href="2021.naacl-demos.1.mp4"/>
+      <pwccode url="https://github.com/VinAIResearch/PhoNLP" additional="false">VinAIResearch/PhoNLP</pwccode>
     </paper>
     <paper id="2">
       <title>Machine-Assisted Script Curation</title>
@@ -7359,8 +7361,8 @@
       <attachment type="Supplementary" hash="32713ecb">2021.naacl-demos.2.Supplementary.zip</attachment>
       <doi>10.18653/v1/2021.naacl-demos.2</doi>
       <bibkey>ciosici-etal-2021-machine</bibkey>
-      <pwccode url="https://github.com/isi-vista/MASC" additional="false">isi-vista/MASC</pwccode>
       <video href="2021.naacl-demos.2.mp4"/>
+      <pwccode url="https://github.com/isi-vista/MASC" additional="false">isi-vista/MASC</pwccode>
     </paper>
     <paper id="3">
       <title><fixed-case>NAMER</fixed-case>: A Node-Based Multitasking Framework for Multi-Hop Knowledge Base Question Answering</title>
@@ -7390,8 +7392,8 @@
       <attachment type="Supplementary" hash="a04bdb7a">2021.naacl-demos.4.Supplementary.mp4</attachment>
       <doi>10.18653/v1/2021.naacl-demos.4</doi>
       <bibkey>ghazarian-etal-2021-discol</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/topical-chat">Topical-Chat</pwcdataset>
       <video href="2021.naacl-demos.4.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/topical-chat">Topical-Chat</pwcdataset>
     </paper>
     <paper id="5">
       <title><fixed-case>FITA</fixed-case>nnotator: A Flexible and Intelligent Text Annotation System</title>
@@ -7488,9 +7490,9 @@
       <url hash="18a45b63">2021.naacl-demos.9</url>
       <doi>10.18653/v1/2021.naacl-demos.9</doi>
       <bibkey>risch-etal-2021-multifaceted</bibkey>
+      <video href="2021.naacl-demos.9.mp4"/>
       <pwccode url="https://github.com/philipphager/faceted-domain-encoder" additional="false">philipphager/faceted-domain-encoder</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/biosses">BIOSSES</pwcdataset>
-      <video href="2021.naacl-demos.9.mp4"/>
     </paper>
     <paper id="10">
       <title>Improving Evidence Retrieval for Automated Explainable Fact-Checking</title>
@@ -7503,13 +7505,13 @@
       <attachment type="Supplementary" hash="6df7b8e1">2021.naacl-demos.10.Supplementary.txt</attachment>
       <doi>10.18653/v1/2021.naacl-demos.10</doi>
       <bibkey>samarinas-etal-2021-improving</bibkey>
+      <video href="2021.naacl-demos.10.mp4"/>
       <pwccode url="https://github.com/algoprog/Quin" additional="false">algoprog/Quin</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/fever">FEVER</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/ms-marco">MS MARCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/scifact">SciFact</pwcdataset>
-      <video href="2021.naacl-demos.10.mp4"/>
     </paper>
     <paper id="11">
       <title>Interactive Plot Manipulation using Natural Language</title>
@@ -7548,8 +7550,8 @@
       <url hash="4f44e392">2021.naacl-demos.13</url>
       <doi>10.18653/v1/2021.naacl-demos.13</doi>
       <bibkey>newman-griffis-etal-2021-textessence</bibkey>
-      <pwccode url="https://github.com/drgriffis/text-essence" additional="false">drgriffis/text-essence</pwccode>
       <video href="2021.naacl-demos.13.mp4"/>
+      <pwccode url="https://github.com/drgriffis/text-essence" additional="false">drgriffis/text-essence</pwccode>
     </paper>
     <paper id="14">
       <title>Supporting <fixed-case>S</fixed-case>panish Writers using Automated Feedback</title>
@@ -7641,8 +7643,8 @@
       <url hash="41c10f7c">2021.naacl-demos.16</url>
       <doi>10.18653/v1/2021.naacl-demos.16</doi>
       <bibkey>wen-etal-2021-resin</bibkey>
-      <pwccode url="https://github.com/resin-kairos/resin-pipeline-public" additional="false">resin-kairos/resin-pipeline-public</pwccode>
       <video href="2021.naacl-demos.16.mp4"/>
+      <pwccode url="https://github.com/resin-kairos/resin-pipeline-public" additional="false">resin-kairos/resin-pipeline-public</pwccode>
     </paper>
     <paper id="17">
       <title><fixed-case>MUDES</fixed-case>: Multilingual Detection of Offensive Spans</title>
@@ -7653,9 +7655,9 @@
       <url hash="8bbcbf4b">2021.naacl-demos.17</url>
       <doi>10.18653/v1/2021.naacl-demos.17</doi>
       <bibkey>ranasinghe-zampieri-2021-mudes</bibkey>
+      <video href="2021.naacl-demos.17.mp4"/>
       <pwccode url="https://github.com/tharindudr/MUDES" additional="false">tharindudr/MUDES</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/olid">OLID</pwcdataset>
-      <video href="2021.naacl-demos.17.mp4"/>
     </paper>
   </volume>
   <volume id="srw" ingest-date="2021-05-24">
@@ -7684,8 +7686,8 @@
       <url hash="e1e7c102">2021.naacl-srw.1</url>
       <doi>10.18653/v1/2021.naacl-srw.1</doi>
       <bibkey>zouhar-2021-sampling</bibkey>
-      <pwccode url="https://github.com/zouharvi/reference-mt-distill" additional="false">zouharvi/reference-mt-distill</pwccode>
       <video href="2021.naacl-srw.1.mp4"/>
+      <pwccode url="https://github.com/zouharvi/reference-mt-distill" additional="false">zouharvi/reference-mt-distill</pwccode>
     </paper>
     <paper id="2">
       <title><fixed-case>I</fixed-case>ce<fixed-case>S</fixed-case>um: An <fixed-case>I</fixed-case>celandic Text Summarization Corpus</title>
@@ -7740,9 +7742,9 @@
       <url hash="05d13951">2021.naacl-srw.6</url>
       <doi>10.18653/v1/2021.naacl-srw.6</doi>
       <bibkey>paun-2021-parallel</bibkey>
+      <video href="2021.naacl-srw.6.mp4"/>
       <pwccode url="https://github.com/stefanpaun/massalign" additional="false">stefanpaun/massalign</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/newsela">Newsela</pwcdataset>
-      <video href="2021.naacl-srw.6.mp4"/>
     </paper>
     <paper id="7">
       <title>Syntax-Based Attention Masking for Neural Machine Translation</title>
@@ -7781,9 +7783,9 @@
       <url hash="613a2a28">2021.naacl-srw.9</url>
       <doi>10.18653/v1/2021.naacl-srw.9</doi>
       <bibkey>xiao-etal-2021-open</bibkey>
+      <video href="2021.naacl-srw.9.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
-      <video href="2021.naacl-srw.9.mp4"/>
     </paper>
     <paper id="10">
       <title>A Sliding-Window Approach to Automatic Creation of Meeting Minutes</title>
@@ -7796,8 +7798,8 @@
       <url hash="9665baad">2021.naacl-srw.10</url>
       <doi>10.18653/v1/2021.naacl-srw.10</doi>
       <bibkey>koay-etal-2021-sliding</bibkey>
-      <pwccode url="https://github.com/ucfnlp/meeting-sliding-window" additional="false">ucfnlp/meeting-sliding-window</pwccode>
       <video href="2021.naacl-srw.10.mp4"/>
+      <pwccode url="https://github.com/ucfnlp/meeting-sliding-window" additional="false">ucfnlp/meeting-sliding-window</pwccode>
     </paper>
     <paper id="11">
       <title>Exploration and Discovery of the <fixed-case>COVID</fixed-case>-19 Literature through Semantic Visualization</title>
@@ -7812,8 +7814,8 @@
       <bibkey>tu-etal-2021-exploration</bibkey>
       <revision id="1" href="2021.naacl-srw.11v1" hash="964dc194"/>
       <revision id="2" href="2021.naacl-srw.11v2" hash="2e0108f5" date="2021-06-28">Added a sponsor to Acknowledgments section</revision>
-      <pwcdataset url="https://paperswithcode.com/dataset/cord-19">CORD-19</pwcdataset>
       <video href="2021.naacl-srw.11.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cord-19">CORD-19</pwcdataset>
     </paper>
     <paper id="12">
       <title>Shuffled-token Detection for Refining Pre-trained <fixed-case>R</fixed-case>o<fixed-case>BERT</fixed-case>a</title>
@@ -7837,8 +7839,8 @@
       <url hash="7eaa71ac">2021.naacl-srw.13</url>
       <doi>10.18653/v1/2021.naacl-srw.13</doi>
       <bibkey>krishnan-ragavan-2021-morphology</bibkey>
-      <pwccode url="https://github.com/arjun-sai-krishnan/tamil-morpho-embeddings" additional="false">arjun-sai-krishnan/tamil-morpho-embeddings</pwccode>
       <video href="2021.naacl-srw.13.mp4"/>
+      <pwccode url="https://github.com/arjun-sai-krishnan/tamil-morpho-embeddings" additional="false">arjun-sai-krishnan/tamil-morpho-embeddings</pwccode>
     </paper>
     <paper id="14">
       <title>Seed Word Selection for Weakly-Supervised Text Classification with Unsupervised Error Estimation</title>
@@ -7850,8 +7852,8 @@
       <url hash="8fd4bbbb">2021.naacl-srw.14</url>
       <doi>10.18653/v1/2021.naacl-srw.14</doi>
       <bibkey>jin-etal-2021-seed</bibkey>
-      <pwccode url="https://github.com/YipingNUS/OptimSeed" additional="false">YipingNUS/OptimSeed</pwccode>
       <video href="2021.naacl-srw.14.mp4"/>
+      <pwccode url="https://github.com/YipingNUS/OptimSeed" additional="false">YipingNUS/OptimSeed</pwccode>
     </paper>
     <paper id="15">
       <title>Multi-Task Learning of Generation and Classification for Emotion-Aware Dialogue Response Generation</title>
@@ -7862,9 +7864,9 @@
       <url hash="3ca106ae">2021.naacl-srw.15</url>
       <doi>10.18653/v1/2021.naacl-srw.15</doi>
       <bibkey>ide-kawahara-2021-multi</bibkey>
+      <video href="2021.naacl-srw.15.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/dailydialog">DailyDialog</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
-      <video href="2021.naacl-srw.15.mp4"/>
     </paper>
     <paper id="16">
       <title>Comparison of Grammatical Error Correction Using Back-Translation Models</title>
@@ -7877,8 +7879,8 @@
       <url hash="b1a75da0">2021.naacl-srw.16</url>
       <doi>10.18653/v1/2021.naacl-srw.16</doi>
       <bibkey>koyama-etal-2021-comparison</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/jfleg">JFLEG</pwcdataset>
       <video href="2021.naacl-srw.16.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/jfleg">JFLEG</pwcdataset>
     </paper>
     <paper id="17">
       <title>Parallel sentences mining with transfer learning in an unsupervised setting</title>
@@ -7918,8 +7920,8 @@
       <url hash="15fc2e51">2021.naacl-srw.19</url>
       <doi>10.18653/v1/2021.naacl-srw.19</doi>
       <bibkey>das-etal-2021-emotion</bibkey>
-      <pwccode url="https://github.com/sagorbrur/bangla-bert" additional="true">sagorbrur/bangla-bert</pwccode>
       <video href="2021.naacl-srw.19.mp4"/>
+      <pwccode url="https://github.com/sagorbrur/bangla-bert" additional="true">sagorbrur/bangla-bert</pwccode>
     </paper>
     <paper id="20">
       <title>Hie-<fixed-case>BART</fixed-case>: Document Summarization with Hierarchical <fixed-case>BART</fixed-case></title>
@@ -7931,8 +7933,8 @@
       <url hash="0d561395">2021.naacl-srw.20</url>
       <doi>10.18653/v1/2021.naacl-srw.20</doi>
       <bibkey>akiyama-etal-2021-hie</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
       <video href="2021.naacl-srw.20.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cnn-daily-mail-1">CNN/Daily Mail</pwcdataset>
     </paper>
     <paper id="21">
       <title>Towards Multi-Modal Text-Image Retrieval to improve Human Reading</title>
@@ -7943,11 +7945,11 @@
       <abstract>In primary school, children’s books, as well as in modern language learning apps, multi-modal learning strategies like illustrations of terms and phrases are used to support reading comprehension. Also, several studies in educational psychology suggest that integrating cross-modal information will improve reading comprehension. We claim that state-of- he-art multi-modal transformers, which could be used in a language learner context to improve human reading, will perform poorly because of the short and relatively simple textual data those models are trained with. To prove our hypotheses, we collected a new multi-modal image-retrieval dataset based on data from Wikipedia. In an in-depth data analysis, we highlight the differences between our dataset and other popular datasets. Additionally, we evaluate several state-of-the-art multi-modal transformers on text-image retrieval on our dataset and analyze their meager results, which verify our claims.</abstract>
       <url hash="d8aae487">2021.naacl-srw.21</url>
       <bibkey>schneider-etal-2021-towards</bibkey>
+      <video href="2021.naacl-srw.21.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/coco">COCO</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptual-captions">Conceptual Captions</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/flickr30k">Flickr30k</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikicaps">WikiCaps</pwcdataset>
-      <video href="2021.naacl-srw.21.mp4"/>
     </paper>
   </volume>
   <volume id="tutorials" ingest-date="2021-05-24">
@@ -7976,6 +7978,7 @@
       <url hash="5cc45700">2021.naacl-tutorials.1</url>
       <doi>10.18653/v1/2021.naacl-tutorials.1</doi>
       <bibkey>yates-etal-2021-pretrained</bibkey>
+      <video href="2021.naacl-tutorials.1.mp4"/>
       <pwcdataset url="https://paperswithcode.com/dataset/asnq">ASNQ</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/beir">BEIR</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/c4">C4</pwcdataset>
@@ -7988,7 +7991,6 @@
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/triviaqa">TriviaQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/webquestions">WebQuestions</pwcdataset>
-      <video href="2021.naacl-tutorials.1.mp4"/>
     </paper>
     <paper id="2">
       <title>Fine-grained Interpretation and Causation Analysis in Deep <fixed-case>NLP</fixed-case> Models</title>
@@ -8136,8 +8138,8 @@
       <url hash="ffdf16fe">2021.naacl-industry.4</url>
       <doi>10.18653/v1/2021.naacl-industry.4</doi>
       <bibkey>shang-etal-2021-entity</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
       <video href="2021.naacl-industry.4.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/wizard-of-wikipedia">Wizard of Wikipedia</pwcdataset>
     </paper>
     <paper id="5">
       <title>Pretrain-Finetune Based Training of Task-Oriented Dialogue Systems in a Real-World Setting</title>
@@ -8224,8 +8226,8 @@
       <url hash="5bdaa941">2021.naacl-industry.10</url>
       <doi>10.18653/v1/2021.naacl-industry.10</doi>
       <bibkey>tyagi-etal-2021-proteno</bibkey>
-      <pwccode url="https://github.com/amazon-research/proteno" additional="false">amazon-research/proteno</pwccode>
       <video href="2021.naacl-industry.10.mp4"/>
+      <pwccode url="https://github.com/amazon-research/proteno" additional="false">amazon-research/proteno</pwccode>
     </paper>
     <paper id="11">
       <title>Addressing the Vulnerability of <fixed-case>NMT</fixed-case> in Input Perturbations</title>
@@ -8239,9 +8241,9 @@
       <url hash="7ea7a894">2021.naacl-industry.11</url>
       <doi>10.18653/v1/2021.naacl-industry.11</doi>
       <bibkey>xu-etal-2021-addressing</bibkey>
+      <video href="2021.naacl-industry.11.mp4"/>
       <pwccode url="https://github.com/wwxu21/CER-MT" additional="false">wwxu21/CER-MT</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/mtnt">MTNT</pwcdataset>
-      <video href="2021.naacl-industry.11.mp4"/>
     </paper>
     <paper id="12">
       <title>Cross-lingual Supervision Improves Unsupervised Neural Machine Translation</title>
@@ -8295,8 +8297,8 @@
       <url hash="288e7385">2021.naacl-industry.15</url>
       <doi>10.18653/v1/2021.naacl-industry.15</doi>
       <bibkey>wang-etal-2021-lightseq</bibkey>
-      <pwccode url="https://github.com/bytedance/lightseq" additional="false">bytedance/lightseq</pwccode>
       <video href="2021.naacl-industry.15.mp4"/>
+      <pwccode url="https://github.com/bytedance/lightseq" additional="false">bytedance/lightseq</pwccode>
     </paper>
     <paper id="16">
       <title>Practical Transformer-based Multilingual Text Classification</title>
@@ -8307,8 +8309,8 @@
       <url hash="f093073c">2021.naacl-industry.16</url>
       <doi>10.18653/v1/2021.naacl-industry.16</doi>
       <bibkey>wang-banko-2021-practical</bibkey>
-      <pwccode url="https://github.com/sentropytechnologies/hateval2019-relabeled" additional="false">sentropytechnologies/hateval2019-relabeled</pwccode>
       <video href="2021.naacl-industry.16.mp4"/>
+      <pwccode url="https://github.com/sentropytechnologies/hateval2019-relabeled" additional="false">sentropytechnologies/hateval2019-relabeled</pwccode>
     </paper>
     <paper id="17">
       <title>An Emotional Comfort Framework for Improving User Satisfaction in <fixed-case>E</fixed-case>-Commerce Customer Service Chatbots</title>
@@ -8368,8 +8370,8 @@
       <award>Best Industry Paper</award>
       <doi>10.18653/v1/2021.naacl-industry.20</doi>
       <bibkey>bianchi-etal-2021-query2prod2vec</bibkey>
-      <pwccode url="https://github.com/coveooss/ecommerce-query-embeddings" additional="false">coveooss/ecommerce-query-embeddings</pwccode>
       <video href="2021.naacl-industry.20.mp4"/>
+      <pwccode url="https://github.com/coveooss/ecommerce-query-embeddings" additional="false">coveooss/ecommerce-query-embeddings</pwccode>
     </paper>
     <paper id="21">
       <title>An Architecture for Accelerated Large-Scale Inference of Transformer-Based Language Models</title>
@@ -8410,9 +8412,9 @@
       <url hash="aa975dcc">2021.naacl-industry.23</url>
       <doi>10.18653/v1/2021.naacl-industry.23</doi>
       <bibkey>yu-etal-2021-technical</bibkey>
+      <video href="2021.naacl-industry.23.mp4"/>
       <pwccode url="https://github.com/wyu97/TransTD" additional="false">wyu97/TransTD</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/techqa">TechQA</pwcdataset>
-      <video href="2021.naacl-industry.23.mp4"/>
     </paper>
     <paper id="24">
       <title>Cost-effective Deployment of <fixed-case>BERT</fixed-case> Models in Serverless Environment</title>
@@ -8484,8 +8486,8 @@
       <url hash="3e7d08e7">2021.naacl-industry.27</url>
       <doi>10.18653/v1/2021.naacl-industry.27</doi>
       <bibkey>lester-etal-2021-intent</bibkey>
-      <pwccode url="https://github.com/blester125/global-local-model" additional="false">blester125/global-local-model</pwccode>
       <video href="2021.naacl-industry.27.mp4"/>
+      <pwccode url="https://github.com/blester125/global-local-model" additional="false">blester125/global-local-model</pwccode>
     </paper>
     <paper id="28">
       <title>Development of an Enterprise-Grade Contract Understanding System</title>
@@ -8540,8 +8542,8 @@
       <url hash="3a5163ba">2021.naacl-industry.30</url>
       <doi>10.18653/v1/2021.naacl-industry.30</doi>
       <bibkey>marek-etal-2021-oodgan</bibkey>
-      <pwcdataset url="https://paperswithcode.com/dataset/rostd">ROSTD</pwcdataset>
       <video href="2021.naacl-industry.30.mp4"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/rostd">ROSTD</pwcdataset>
     </paper>
     <paper id="31">
       <title>Coherent and Concise Radiology Report Generation via Context Specific Image Representations and Orthogonal Sentence States</title>
@@ -8641,10 +8643,10 @@
       <url hash="b748118c">2021.naacl-industry.38</url>
       <doi>10.18653/v1/2021.naacl-industry.38</doi>
       <bibkey>qi-etal-2021-benchmarking</bibkey>
+      <video href="2021.naacl-industry.38.mp4"/>
       <pwccode url="https://github.com/haodeqi/BenchmarkingIntentDetection" additional="false">haodeqi/BenchmarkingIntentDetection</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/clinc150">CLINC150</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hint3">HINT3</pwcdataset>
-      <video href="2021.naacl-industry.38.mp4"/>
     </paper>
     <paper id="39">
       <title>Industry Scale Semi-Supervised Learning for Natural Language Understanding</title>
diff --git a/data/xml/2022.acl.xml b/data/xml/2022.acl.xml
index ada0f67320..16c9118c01 100644
--- a/data/xml/2022.acl.xml
+++ b/data/xml/2022.acl.xml
@@ -93,6 +93,7 @@
       <doi>10.18653/v1/2022.acl-long.5</doi>
       <pwcdataset url="https://paperswithcode.com/dataset/drop">DROP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/hybridqa">HybridQA</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/tat-qa">TAT-QA</pwcdataset>
     </paper>
     <paper id="6">
       <title>Domain Adaptation in Multilingual and Multi-Domain Monolingual Settings for Complex Word Identification</title>
@@ -1250,6 +1251,7 @@
       <doi>10.18653/v1/2022.acl-long.78</doi>
       <pwccode url="https://github.com/microsoft/hitab" additional="false">microsoft/hitab</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/finqa">FinQA</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/tat-qa">TAT-QA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/totto">ToTTo</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/wikisql">WikiSQL</pwcdataset>
     </paper>
@@ -2814,7 +2816,7 @@
       <attachment type="software" hash="b432ecaf">2022.acl-long.182.software.zip</attachment>
       <bibkey>lin-etal-2022-roles</bibkey>
       <doi>10.18653/v1/2022.acl-long.182</doi>
-      <pwccode url="https://github.com/xiaolinandy/rods" additional="false">xiaolinandy/rods</pwccode>
+      <pwccode url="https://github.com/atulkum/pointer_summarizer" additional="true">atulkum/pointer_summarizer</pwccode>
     </paper>
     <paper id="183">
       <title><fixed-case>C</fixed-case>lar<fixed-case>ET</fixed-case>: Pre-training a Correlation-Aware Context-To-Event Transformer for Event-Centric Generation and Classification</title>
@@ -3909,6 +3911,7 @@ in the Case of Unambiguous Gender</title>
       <attachment type="software" hash="53624271">2022.acl-long.252.software.zip</attachment>
       <bibkey>dankers-etal-2022-transformer</bibkey>
       <doi>10.18653/v1/2022.acl-long.252</doi>
+      <pwccode url="https://github.com/vernadankers/mt_idioms" additional="false">vernadankers/mt_idioms</pwccode>
     </paper>
     <paper id="253">
       <title><fixed-case>C</fixed-case>onditional<fixed-case>QA</fixed-case>: A Complex Reading Comprehension Dataset with Conditional Answers</title>
@@ -4524,6 +4527,7 @@ in the Case of Unambiguous Gender</title>
       <attachment type="software" hash="4f304282">2022.acl-long.294.software.zip</attachment>
       <bibkey>yu-etal-2022-rst</bibkey>
       <doi>10.18653/v1/2022.acl-long.294</doi>
+      <pwccode url="https://github.com/yunan4nlp/e-nnrstparser" additional="false">yunan4nlp/e-nnrstparser</pwccode>
     </paper>
     <paper id="295">
       <title><fixed-case>S</fixed-case>im<fixed-case>KGC</fixed-case>: Simple Contrastive Knowledge Graph Completion with Pre-trained Language Models</title>
@@ -7029,6 +7033,7 @@ in the Case of Unambiguous Gender</title>
       <pwcdataset url="https://paperswithcode.com/dataset/hybridqa">HybridQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/math">MATH</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mathqa">MathQA</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/tat-qa">TAT-QA</pwcdataset>
     </paper>
     <paper id="455">
       <title>Transformers in the loop: Polarity in neural models of language</title>
@@ -9754,6 +9759,7 @@ in the Case of Unambiguous Gender</title>
       <url hash="b3db0099">2022.acl-short.30</url>
       <bibkey>weller-etal-2022-use</bibkey>
       <doi>10.18653/v1/2022.acl-short.30</doi>
+      <pwccode url="https://github.com/orionw/mtlvsift" additional="false">orionw/mtlvsift</pwccode>
     </paper>
     <paper id="31">
       <title>Leveraging Explicit Lexico-logical Alignments in Text-to-<fixed-case>SQL</fixed-case> Parsing</title>
diff --git a/data/xml/2022.cl.xml b/data/xml/2022.cl.xml
index 688c143d78..66a2087e17 100644
--- a/data/xml/2022.cl.xml
+++ b/data/xml/2022.cl.xml
@@ -38,6 +38,8 @@
       <pages>77–117</pages>
       <url hash="3a678836">2022.cl-1.3</url>
       <bibkey>ghosal-etal-2022-novelty</bibkey>
+      <pwcdataset url="https://paperswithcode.com/dataset/multinli">MultiNLI</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
     </paper>
     <paper id="4">
       <title>Improved N-Best Extraction with an Evaluation on Language Data</title>
@@ -77,6 +79,8 @@
       <pages>155–205</pages>
       <url hash="ad5beda4">2022.cl-1.6</url>
       <bibkey>jin-etal-2022-deep</bibkey>
+      <pwccode url="https://github.com/fuzhenxin/Style-Transfer-in-Text" additional="true">fuzhenxin/Style-Transfer-in-Text</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/gyafc">GYAFC</pwcdataset>
     </paper>
     <paper id="7">
       <title>Probing Classifiers: Promises, Shortcomings, and Advances</title>
@@ -96,6 +100,9 @@
       <pages>221–232</pages>
       <url hash="d99b7b9c">2022.cl-1.8</url>
       <bibkey>faruqui-hakkani-tur-2022-revisiting</bibkey>
+      <pwcdataset url="https://paperswithcode.com/dataset/coqa">CoQA</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/squad">SQuAD</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/spoken-squad">Spoken-SQuAD</pwcdataset>
     </paper>
     <paper id="9">
       <title>Natural Language Processing: A Machine Learning Perspective by <fixed-case>Y</fixed-case>ue <fixed-case>Z</fixed-case>hang and Zhiyang Teng</title>
diff --git a/data/xml/2022.deelio.xml b/data/xml/2022.deelio.xml
index 44120faa5b..1e40e538df 100644
--- a/data/xml/2022.deelio.xml
+++ b/data/xml/2022.deelio.xml
@@ -95,6 +95,7 @@
       <url hash="2993c9a8">2022.deelio-1.6</url>
       <bibkey>nararatwong-etal-2022-kiqa</bibkey>
       <doi>10.18653/v1/2022.deelio-1.6</doi>
+      <pwcdataset url="https://paperswithcode.com/dataset/tat-qa">TAT-QA</pwcdataset>
     </paper>
     <paper id="7">
       <title>Trans-<fixed-case>KBLSTM</fixed-case>: An External Knowledge Enhanced Transformer <fixed-case>B</fixed-case>i<fixed-case>LSTM</fixed-case> Model for Tabular Reasoning</title>
diff --git a/data/xml/2022.ecnlp.xml b/data/xml/2022.ecnlp.xml
index 85d9ca33d8..f6b0169dca 100644
--- a/data/xml/2022.ecnlp.xml
+++ b/data/xml/2022.ecnlp.xml
@@ -110,6 +110,7 @@
       <bibkey>jain-etal-2022-comparative</bibkey>
       <doi>10.18653/v1/2022.ecnlp-1.7</doi>
       <pwccode url="https://github.com/wing-nus/comparative-snippet-generation-dataset" additional="false">wing-nus/comparative-snippet-generation-dataset</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/spot">SPOT</pwcdataset>
     </paper>
     <paper id="8">
       <title>Textual Content Moderation in <fixed-case>C</fixed-case>2<fixed-case>C</fixed-case> Marketplace</title>
diff --git a/data/xml/2022.findings.xml b/data/xml/2022.findings.xml
index 589102cbde..0a4887085a 100644
--- a/data/xml/2022.findings.xml
+++ b/data/xml/2022.findings.xml
@@ -1622,6 +1622,7 @@
       <url hash="e140b04a">2022.findings-acl.107</url>
       <bibkey>li-etal-2022-c3kg</bibkey>
       <doi>10.18653/v1/2022.findings-acl.107</doi>
+      <pwccode url="https://github.com/xiaomi/c3kg" additional="false">xiaomi/c3kg</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/atomic">ATOMIC</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conceptnet">ConceptNet</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mod-1">MOD</pwcdataset>
@@ -4278,6 +4279,8 @@
       <attachment type="software" hash="0a86dadd">2022.findings-acl.284.software.zip</attachment>
       <bibkey>zhang-etal-2022-improving</bibkey>
       <doi>10.18653/v1/2022.findings-acl.284</doi>
+      <pwccode url="https://github.com/zhangcen456/ib" additional="false">zhangcen456/ib</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/imdb-movie-reviews">IMDb Movie Reviews</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/sst">SST</pwcdataset>
     </paper>
diff --git a/data/xml/D14.xml b/data/xml/D14.xml
index 89672e0ce6..d06aa1554a 100644
--- a/data/xml/D14.xml
+++ b/data/xml/D14.xml
@@ -1922,6 +1922,7 @@
       <doi>10.3115/v1/D14-1181</doi>
       <bibkey>kim-2014-convolutional</bibkey>
       <pwccode url="" additional="true"/>
+      <pwcdataset url="https://paperswithcode.com/dataset/cped">CPED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/glue">GLUE</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/mpqa-opinion-corpus">MPQA Opinion Corpus</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/snli">SNLI</pwcdataset>
diff --git a/data/xml/D19.xml b/data/xml/D19.xml
index 891a1a59bf..714e9df3d5 100644
--- a/data/xml/D19.xml
+++ b/data/xml/D19.xml
@@ -225,6 +225,7 @@
       <doi>10.18653/v1/D19-1015</doi>
       <bibkey>ghosal-etal-2019-dialoguegcn</bibkey>
       <pwccode url="https://github.com/SenticNet/conv-emotion" additional="true">SenticNet/conv-emotion</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/cped">CPED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/semaine">SEMAINE</pwcdataset>
diff --git a/data/xml/E17.xml b/data/xml/E17.xml
index 0ea16da13a..ae0eb61983 100644
--- a/data/xml/E17.xml
+++ b/data/xml/E17.xml
@@ -2065,6 +2065,7 @@
       <bibkey>joulin-etal-2017-bag</bibkey>
       <pwccode url="https://github.com/facebookresearch/fastText" additional="true">facebookresearch/fastText</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/ag-news">AG News</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/cped">CPED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dbpedia">DBpedia</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/yfcc100m">YFCC100M</pwcdataset>
     </paper>
diff --git a/data/xml/N19.xml b/data/xml/N19.xml
index ca26ea34b8..f613dcf06b 100644
--- a/data/xml/N19.xml
+++ b/data/xml/N19.xml
@@ -5752,6 +5752,7 @@
       <bibkey>devlin-etal-2019-bert</bibkey>
       <pwccode url="https://github.com/google-research/bert" additional="true">google-research/bert</pwccode>
       <pwcdataset url="https://paperswithcode.com/dataset/bc5cdr">BC5CDR</pwcdataset>
+      <pwcdataset url="https://paperswithcode.com/dataset/cped">CPED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/conll-2003">CoNLL-2003</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/coqa">CoQA</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/dbpedia">DBpedia</pwcdataset>
diff --git a/data/xml/P17.xml b/data/xml/P17.xml
index b2d5720cb0..7e317ff5a4 100644
--- a/data/xml/P17.xml
+++ b/data/xml/P17.xml
@@ -1169,6 +1169,7 @@ two word-vectors results in a vector that is only a small angle away from the ve
       <video href="https://vimeo.com/234955981"/>
       <bibkey>poria-etal-2017-context</bibkey>
       <pwccode url="https://github.com/senticnet/sc-lstm" additional="true">senticnet/sc-lstm</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/cped">CPED</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/iemocap">IEMOCAP</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/meld">MELD</pwcdataset>
       <pwcdataset url="https://paperswithcode.com/dataset/multimodal-opinionlevel-sentiment-intensity">Multimodal Opinionlevel Sentiment Intensity</pwcdataset>
diff --git a/data/xml/W19.xml b/data/xml/W19.xml
index 292726c9aa..00696ba060 100644
--- a/data/xml/W19.xml
+++ b/data/xml/W19.xml
@@ -14499,6 +14499,7 @@ One of the references was wrong therefore it is corrected to cite the appropriat
       <url hash="71f29b16">W19-6122</url>
       <bibkey>lillie-etal-2019-joint</bibkey>
       <pwccode url="https://github.com/danish-stance-detectors/RumourResolution" additional="true">danish-stance-detectors/RumourResolution</pwccode>
+      <pwcdataset url="https://paperswithcode.com/dataset/dast">DAST</pwcdataset>
     </paper>
     <paper id="23">
       <title>Named-Entity Recognition for <fixed-case>N</fixed-case>orwegian</title>