Rebuilt site

llmrisks · Nov 15, 2023 · 1040a8a · 1040a8a
1 parent a9de020
commit 1040a8a
Show file tree

Hide file tree

Showing 19 changed files with 150 additions and 144 deletions.
diff --git a/docs/shneiderman.pdf b/docs/shneiderman.pdf
diff --git a/index.html b/index.html
diff --git a/index.xml b/index.xml
diff --git a/post/index.html b/post/index.html
@@ -109,7 +109,8 @@ <h2><a href="/week10/">Week 10: Data Selection for LLMs</a></h2>
 
 
 (see bottom for assigned readings and questions)
-Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim, Stephanie Schoch Blogging Team: Aparna Kishore, Elena Long, Erzhen Hu, Jingping Wan
+Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim, Stephanie Schoch 
+Blogging Team: Aparna Kishore, Elena Long, Erzhen Hu, Jingping Wan
 Monday, 30 October: Data Selection for Fine-tuning LLMs Question: Would more models help? We&rsquo;ve discussed so many risks and issues of GenAI so far and one question is that it can be difficult for us to come up with a possible solution to these problems.
 <p class="text-right"><a href="/week10/">Read More…</a></p>
 
@@ -127,7 +128,7 @@ <h2><a href="/week9/">Week 9: Interpretability</a></h2>
 (see bottom for assigned readings and questions)
 Presenting Team: Anshuman Suri, Jacob Christopher, Kasra Lekan, Kaylee Liu, My Dinh
 Blogging Team: Hamza Khalid, Liu Zhe, Peng Wang, Sikun Guo, Yinhan He, Zhepei Wei
-Monday, 23 October: Interpretability: Overview, Limitations, &amp; Challenges Definition of Interpretability Interpretability in the context of artificial intelligence (AI) and machine learning refers to the extent to which a model&rsquo;s decisions, predictions, or internal workings can be understood and explained by humans.
+Monday, 23 October: Interpretability: Overview, Limitations, &amp; Challenges Definition of Interpretability  Interpretability in the context of artificial intelligence (AI) and machine learning refers to the extent to which a model&rsquo;s decisions, predictions, or internal workings can be understood and explained by humans.
 <p class="text-right"><a href="/week9/">Read More…</a></p>
 
 
@@ -159,8 +160,10 @@ <h2><a href="/week7/">Week 7: GANs and DeepFakes</a></h2>
 
 
 (see bottom for assigned readings and questions)
-Presenting Team: Aparna Kishore, Elena Long, Erzhen Hu, Jingping Wan Blogging Team: Haochen Liu, Haolin Liu, Ji Hyun Kim, Stephanie Schoch, Xueren Ge Monday, 9 October: Generative Adversarial Networks and DeepFakes Today's topic is how to utilize generative adversarial networks to create fake images and how to identify the images generated by these models.
-Generative Adversarial Network (GAN) is a revolutionary deep learning framework that pits two neural networks against each other in a creative showdown.
+Presenting Team: Aparna Kishore, Elena Long, Erzhen Hu, Jingping Wan 
+Blogging Team: Haochen Liu, Haolin Liu, Ji Hyun Kim, Stephanie Schoch, Xueren Ge 
+Monday, 9 October: Generative Adversarial Networks and DeepFakes Today's topic is how to utilize generative adversarial networks to create fake images and how to identify the images generated by these models.
+ Generative Adversarial Network (GAN) is a revolutionary deep learning framework that pits two neural networks against each other in a creative showdown.
 <p class="text-right"><a href="/week7/">Read More…</a></p>
 
 
@@ -177,7 +180,7 @@ <h2><a href="/week5/">Week 5: Hallucination</a></h2>
 (see bottom for assigned readings and questions)
 Hallucination (Week 5) Presenting Team: Liu Zhe, Peng Wang, Sikun Guo, Yinhan He, Zhepei Wei
 Blogging Team: Anshuman Suri, Jacob Christopher, Kasra Lekan, Kaylee Liu, My Dinh
-Wednesday, September 27th: Intro to Hallucination People Hallucinate Too Hallucination Definition There are three types of hallucinations according to the “Siren's Song in the AI Ocean” paper: Input-conflict: This subcategory of hallucinations deviates from user input. Context-conflict: Context-conflict hallucinations occur when a model generates contradicting information within a response.
+Wednesday, September 27th: Intro to Hallucination      People Hallucinate Too         Hallucination Definition  There are three types of hallucinations according to the “Siren's Song in the AI Ocean” paper:  Input-conflict: This subcategory of hallucinations deviates from user input.
 <p class="text-right"><a href="/week5/">Read More…</a></p>
 
 
@@ -209,8 +212,9 @@ <h2><a href="/week3/">Week 3: Prompting and Bias</a></h2>
 
 
 (see bottom for assigned readings and questions)
-Prompt Engineering (Week 3) Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim, Stephanie Schoch Blogging Team: Aparna Kishore, Erzhen Hu, Elena Long, Jingping Wan
-(Monday, 09/11/2023) Prompt Engineering Warm-up questions What is Prompt Engineering? How is prompt-based learning different from traditional supervised learning? In-context learning and different types of prompts What is the difference between prompts and fine-tuning? When is the best to use prompts vs fine-tuning?
+Prompt Engineering (Week 3) Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim, Stephanie Schoch 
+Blogging Team: Aparna Kishore, Erzhen Hu, Elena Long, Jingping Wan
+ (Monday, 09/11/2023) Prompt Engineering  Warm-up questions What is Prompt Engineering? How is prompt-based learning different from traditional supervised learning? In-context learning and different types of prompts What is the difference between prompts and fine-tuning? When is the best to use prompts vs fine-tuning?
 <p class="text-right"><a href="/week3/">Read More…</a></p>
 
 
@@ -225,7 +229,7 @@ <h2><a href="/week2/">Week 2: Alignment</a></h2>
 
 
 (see bottom for assigned readings and questions)
-Table of Contents (Monday, 09/04/2023) Introduction to Alignment Introduction to AI Alignment and Failure Cases Discussion Questions The Alignment Problem from a Deep Learning Perspective Group of RL-based methods Group of LLM-based methods Group of Other ML methods (Wednesday, 09/06/2023) Alignment Challenges and Solutions Opening Discussion Introduction to Red-Teaming In-class Activity (5 groups) How to use Red-Teaming? Alignment Solutions LLM Jailbreaking - Introduction LLM Jailbreaking - Demo Observations Potential Improvement Ideas Closing Remarks (by Prof.
+Table of Contents  (Monday, 09/04/2023) Introduction to Alignment  Introduction to AI Alignment and Failure Cases  Discussion Questions   The Alignment Problem from a Deep Learning Perspective  Group of RL-based methods Group of LLM-based methods Group of Other ML methods     (Wednesday, 09/06/2023) Alignment Challenges and Solutions  Opening Discussion Introduction to Red-Teaming  In-class Activity (5 groups) How to use Red-Teaming?
 <p class="text-right"><a href="/week2/">Read More…</a></p>
 
 
@@ -241,7 +245,7 @@ <h2><a href="/week1/">Week 1: Introduction</a></h2>
 
 (see bottom for assigned readings and questions)
 Attention, Transformers, and BERT Monday, 28 August
-Transformers1 are a class of deep learning models that have revolutionized the field of natural language processing (NLP) and various other domains. The concept of transformers originated as an attempt to address the limitations of traditional recurrent neural networks (RNNs) in sequential data processing. Here&rsquo;s an overview of transformers&rsquo; evolution and significance.
+Transformers1 are a class of deep learning models that have revolutionized the field of natural language processing (NLP) and various other domains. The concept of transformers originated as an attempt to address the limitations of traditional recurrent neural networks (RNNs) in sequential data processing. Here&rsquo;s an overview of transformers' evolution and significance.
 Background and Origin RNNs2 were one of the earliest models used for sequence-based tasks in machine learning.
 <p class="text-right"><a href="/week1/">Read More…</a></p>
 

diff --git a/post/index.xml b/post/index.xml
@@ -27,7 +27,8 @@ Monday, November 6: Watermarking LLM Output Recent instances of AI-generated tex
       <author>[email protected] (David Evans)</author>
       <guid>https://llmrisks.github.io/week10/</guid>
       <description>(see bottom for assigned readings and questions)
-Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim, Stephanie Schoch Blogging Team: Aparna Kishore, Elena Long, Erzhen Hu, Jingping Wan
+Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim, Stephanie Schoch 
+Blogging Team: Aparna Kishore, Elena Long, Erzhen Hu, Jingping Wan
 Monday, 30 October: Data Selection for Fine-tuning LLMs Question: Would more models help? We&amp;rsquo;ve discussed so many risks and issues of GenAI so far and one question is that it can be difficult for us to come up with a possible solution to these problems.</description>
     </item>
 
@@ -40,7 +41,7 @@ Monday, 30 October: Data Selection for Fine-tuning LLMs Question: Would more mod
       <description>(see bottom for assigned readings and questions)
 Presenting Team: Anshuman Suri, Jacob Christopher, Kasra Lekan, Kaylee Liu, My Dinh
 Blogging Team: Hamza Khalid, Liu Zhe, Peng Wang, Sikun Guo, Yinhan He, Zhepei Wei
-Monday, 23 October: Interpretability: Overview, Limitations, &amp;amp; Challenges Definition of Interpretability Interpretability in the context of artificial intelligence (AI) and machine learning refers to the extent to which a model&amp;rsquo;s decisions, predictions, or internal workings can be understood and explained by humans.</description>
+Monday, 23 October: Interpretability: Overview, Limitations, &amp;amp; Challenges Definition of Interpretability  Interpretability in the context of artificial intelligence (AI) and machine learning refers to the extent to which a model&amp;rsquo;s decisions, predictions, or internal workings can be understood and explained by humans.</description>
     </item>
 
     <item>
@@ -62,8 +63,10 @@ Monday, 16 Oct: Diving into the History of Machine Translation Let&amp;rsquo;s k
       <author>[email protected] (David Evans)</author>
       <guid>https://llmrisks.github.io/week7/</guid>
       <description>(see bottom for assigned readings and questions)
-Presenting Team: Aparna Kishore, Elena Long, Erzhen Hu, Jingping Wan Blogging Team: Haochen Liu, Haolin Liu, Ji Hyun Kim, Stephanie Schoch, Xueren Ge Monday, 9 October: Generative Adversarial Networks and DeepFakes Today&#39;s topic is how to utilize generative adversarial networks to create fake images and how to identify the images generated by these models.
-Generative Adversarial Network (GAN) is a revolutionary deep learning framework that pits two neural networks against each other in a creative showdown.</description>
+Presenting Team: Aparna Kishore, Elena Long, Erzhen Hu, Jingping Wan 
+Blogging Team: Haochen Liu, Haolin Liu, Ji Hyun Kim, Stephanie Schoch, Xueren Ge 
+Monday, 9 October: Generative Adversarial Networks and DeepFakes Today&#39;s topic is how to utilize generative adversarial networks to create fake images and how to identify the images generated by these models.
+ Generative Adversarial Network (GAN) is a revolutionary deep learning framework that pits two neural networks against each other in a creative showdown.</description>
     </item>
 
     <item>
@@ -75,7 +78,7 @@ Generative Adversarial Network (GAN) is a revolutionary deep learning framework
       <description>(see bottom for assigned readings and questions)
 Hallucination (Week 5) Presenting Team: Liu Zhe, Peng Wang, Sikun Guo, Yinhan He, Zhepei Wei
 Blogging Team: Anshuman Suri, Jacob Christopher, Kasra Lekan, Kaylee Liu, My Dinh
-Wednesday, September 27th: Intro to Hallucination People Hallucinate Too Hallucination Definition There are three types of hallucinations according to the “Siren&#39;s Song in the AI Ocean” paper: Input-conflict: This subcategory of hallucinations deviates from user input. Context-conflict: Context-conflict hallucinations occur when a model generates contradicting information within a response.</description>
+Wednesday, September 27th: Intro to Hallucination      People Hallucinate Too         Hallucination Definition  There are three types of hallucinations according to the “Siren&#39;s Song in the AI Ocean” paper:  Input-conflict: This subcategory of hallucinations deviates from user input.</description>
     </item>
 
     <item>
@@ -97,8 +100,9 @@ Monday, September 18 Jingfeng Yang, Hongye Jin, Ruixiang Tang, Xiaotian Han, Qiz
       <author>[email protected] (David Evans)</author>
       <guid>https://llmrisks.github.io/week3/</guid>
       <description>(see bottom for assigned readings and questions)
-Prompt Engineering (Week 3) Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim, Stephanie Schoch Blogging Team: Aparna Kishore, Erzhen Hu, Elena Long, Jingping Wan
-(Monday, 09/11/2023) Prompt Engineering Warm-up questions What is Prompt Engineering? How is prompt-based learning different from traditional supervised learning? In-context learning and different types of prompts What is the difference between prompts and fine-tuning? When is the best to use prompts vs fine-tuning?</description>
+Prompt Engineering (Week 3) Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim, Stephanie Schoch 
+Blogging Team: Aparna Kishore, Erzhen Hu, Elena Long, Jingping Wan
+ (Monday, 09/11/2023) Prompt Engineering  Warm-up questions What is Prompt Engineering? How is prompt-based learning different from traditional supervised learning? In-context learning and different types of prompts What is the difference between prompts and fine-tuning? When is the best to use prompts vs fine-tuning?</description>
     </item>
 
     <item>
@@ -108,7 +112,7 @@ Prompt Engineering (Week 3) Presenting Team: Haolin Liu, Xueren Ge, Ji Hyun Kim,
       <author>[email protected] (David Evans)</author>
       <guid>https://llmrisks.github.io/week2/</guid>
       <description>(see bottom for assigned readings and questions)
-Table of Contents (Monday, 09/04/2023) Introduction to Alignment Introduction to AI Alignment and Failure Cases Discussion Questions The Alignment Problem from a Deep Learning Perspective Group of RL-based methods Group of LLM-based methods Group of Other ML methods (Wednesday, 09/06/2023) Alignment Challenges and Solutions Opening Discussion Introduction to Red-Teaming In-class Activity (5 groups) How to use Red-Teaming? Alignment Solutions LLM Jailbreaking - Introduction LLM Jailbreaking - Demo Observations Potential Improvement Ideas Closing Remarks (by Prof.</description>
+Table of Contents  (Monday, 09/04/2023) Introduction to Alignment  Introduction to AI Alignment and Failure Cases  Discussion Questions   The Alignment Problem from a Deep Learning Perspective  Group of RL-based methods Group of LLM-based methods Group of Other ML methods     (Wednesday, 09/06/2023) Alignment Challenges and Solutions  Opening Discussion Introduction to Red-Teaming  In-class Activity (5 groups) How to use Red-Teaming?</description>
     </item>
 
     <item>
@@ -119,7 +123,7 @@ Table of Contents (Monday, 09/04/2023) Introduction to Alignment Introduction to
       <guid>https://llmrisks.github.io/week1/</guid>
       <description>(see bottom for assigned readings and questions)
 Attention, Transformers, and BERT Monday, 28 August
-Transformers1 are a class of deep learning models that have revolutionized the field of natural language processing (NLP) and various other domains. The concept of transformers originated as an attempt to address the limitations of traditional recurrent neural networks (RNNs) in sequential data processing. Here&amp;rsquo;s an overview of transformers&amp;rsquo; evolution and significance.
+Transformers1 are a class of deep learning models that have revolutionized the field of natural language processing (NLP) and various other domains. The concept of transformers originated as an attempt to address the limitations of traditional recurrent neural networks (RNNs) in sequential data processing. Here&amp;rsquo;s an overview of transformers&#39; evolution and significance.
 Background and Origin RNNs2 were one of the earliest models used for sequence-based tasks in machine learning.</description>
     </item>
 
@@ -143,7 +147,7 @@ Once you&amp;rsquo;ve accepted the invitation, you should be able to visit https
       <guid>https://llmrisks.github.io/class0/</guid>
       <description>I&amp;rsquo;ve updated the Schedule and Bi-Weekly Schedule based on the discussions today.
 The plan is below:
-Week Lead Team Blogging Team Everyone Else Two Weeks Before Come up with idea for the week and planned readings, send to me by 5:29pm on Tuesday (2 weeks - 1 day before) - - Week Before Post plan and questions in github discussions by no later than 9am Wednesday; prepare for leading meetings Prepare plan for blogging (how you will divide workload, collaborative tools for taking notes and writing) Read/do materials and respond to preparation questions in github discussions (by 5:29pm Sunday) Week of Leading Meetings Lead interesting, engaging, and illuminating meetings!</description>
+  Week Lead Team Blogging Team Everyone Else     Two Weeks Before  Come up with idea for the week and planned readings, send to me by 5:29pm on Tuesday (2 weeks - 1 day before)  -  -    Week Before  Post plan and questions in github discussions by no later than 9am Wednesday; prepare for leading meetings   Prepare plan for blogging (how you will divide workload, collaborative tools for taking notes and writing)   Read/do materials and respond to preparation questions in github discussions (by 5:29pm Sunday)    Week of Leading Meetings  Lead interesting, engaging, and illuminating meetings!</description>
     </item>
 
     <item>
@@ -153,7 +157,8 @@ Week Lead Team Blogging Team Everyone Else Two Weeks Before Come up with idea fo
       <author>[email protected] (David Evans)</author>
       <guid>https://llmrisks.github.io/updates/</guid>
       <description>Some materials have been posted on the course site:
-Syllabus Schedule (you will find out which team you are on at the first class Wednesday) Readings and Topics (a start on a list of some potential readings and topics that we might want to cover) Dall-E Prompt: &#34;comic style drawing of a phd seminar on AI&#34; </description>
+ Syllabus Schedule (you will find out which team you are on at the first class Wednesday) Readings and Topics (a start on a list of some potential readings and topics that we might want to cover)   
+Dall-E Prompt: &#34;comic style drawing of a phd seminar on AI&#34;  </description>
     </item>
 
     <item>
@@ -163,8 +168,8 @@ Syllabus Schedule (you will find out which team you are on at the first class We
       <author>[email protected] (David Evans)</author>
       <guid>https://llmrisks.github.io/survey/</guid>
       <description>Please submit this welcome survey before 8:59pm on Monday, August 21:
-https://forms.gle/dxhFmJH7WRs32s1ZA
-Your answers won&amp;rsquo;t be shared publicly, but I will use the responses to the survey to plan the seminar, including forming teams, and may share some aggregate and anonymized results and anonymized quotes from the surveys.</description>
+ https://forms.gle/dxhFmJH7WRs32s1ZA
+ Your answers won&amp;rsquo;t be shared publicly, but I will use the responses to the survey to plan the seminar, including forming teams, and may share some aggregate and anonymized results and anonymized quotes from the surveys.</description>
     </item>
 
     <item>
@@ -174,7 +179,7 @@ Your answers won&amp;rsquo;t be shared publicly, but I will use the responses to
       <author>[email protected] (David Evans)</author>
       <guid>https://llmrisks.github.io/welcome/</guid>
       <description>Full Transcript
-Seminar Plan The actual seminar won&amp;rsquo;t be fully planned by GPT-4, but more information on it won&amp;rsquo;t be available until later.
+ Seminar Plan The actual seminar won&amp;rsquo;t be fully planned by GPT-4, but more information on it won&amp;rsquo;t be available until later.
 I&amp;rsquo;m expecting the structure and format to that combines aspects of this seminar on adversarial machine learning and this course on computing ethics, but with a topic focused on learning as much as we can about the potential for both good and harm from generative AI (including large language models) and things we can do (mostly technically, but including policy) to mitigate the harms.</description>
     </item>