- 수정 및 업데이트 요청 : 구혜연([email protected]), 이윤재([email protected]), 민지웅([email protected])
- 최신반 스케쥴: 구글 시트
-
01: Large-scale LM에 대한 얕고 넓은 지식들 (part 1)
- Video, Presentation
- Keywords: Large-scale LM, Parallelism
- Team: 고현웅(발표), 김대웅, 이윤재
-
02: Efficient Transformers: A Survey
- Video, Presentation
- Keywords: Transformer, Scalable, Longer Sequence
- Team: 김백수, 이다인, 조주현
-
03: TOD-BERT Pre-trained Natural Language Understanding for Task-Oriented Dialogue
- Video, Presentation
- Keywords: BERT, Tasl Oriented Doalogue
- Team: 이정훈, 박동주, 이인환, 조소영
-
04: Language ID in the Wild: Unexpected Challenges on the Path to a Thousand-Language Web Text Corpus
- Video, Presentation
- Keywords: Language ID, LangID, Large Text, Data, CCNet
- Team: 박소희(발표), 김민현, 김성환
-
05 : REALM : Retrieval-Augmented Language Model Pre-Training
- Video, Presentation
- Keywords: BERT, Open-QA
- Team: 김대규, 모윤호, 원혜진(발표)
-
06 : How Good is Your Tokenizer? On the Monolingual Performance of Multilingual Language Models
- Video, Presentation
- Keywords: Tokenizer, Multilingual, Monolingual
- Team: 김한민, 이유경(발표), 지승현
-
07 : Survey of Document-level Relation Extraction
- Video, Presentation
- Keywords: Document-level, Relation Extraction
- Team: 이창희, 하헌진, 현지웅(발표)
-
08 : KM-BART: Knowledge Enhanced Multimodal BART for Visual Commonsense Generation
- Video, Presentation
- Keywords: Multimodal, Visual Commonsense Generation, Knowledge-based Commonsense Generation
- Team: 민지웅, 이나연, 임정환
-
09 : SimCSE: Simple Contrastive Learning of Sentence Embeddings
- Video, Presentation
- Keywords: Sentence embedding, Constrative learning, Dropout
- Team: 김한길, 김선행, 옥창원
-
10 : ERNIE 3.0: Large-scale Knowledge Enhanced Pre-training for Language Understanding and Generation
- Video, Presentation
- Keywords: Continual Multi-Paradigms Unified Pre-training Framework, Universal Representation Module, Task Specific Representation Module
- Team: 김은희(발표), 박한솔, 구혜연