Discussions

• 1 You must be logged in to vote

  📆

  [Reading Group]: Information Locality as an Inductive Bias for Neural Language Models

  nilinykh announced Sep 4, 2026 in Meetings & Announcements

  

  

  0
• 1 You must be logged in to vote

  📚

  [Paper Suggestion]: Language Acquisition Device in Large Language Models

  EkaterinaVoloshina started Sep 4, 2026 in Paper Suggestions

  

  

  0
• 3 You must be logged in to vote

  📚

  [Paper Suggestion]: The Emergent Symbolic Structure of Artificial Neural Networks

  moajohansson started Sep 2, 2026 in Paper Suggestions

  

  

  0
• 6 You must be logged in to vote

  📚

  [Paper Suggestion]: How Chain-of-Thought Works? Tracing Information Flow from Decoding, Projection, and Activation

  lucamodica started Aug 28, 2026 in Paper Suggestions

  

  

  0
• 1 You must be logged in to vote

  📆

  [Reading Group]: Position: LLMs can't jump

  nicaudinet announced Aug 21, 2026 in Meetings & Announcements · Closed

  

  

  

  

  3
• 4 You must be logged in to vote

  📚

  [Paper Suggestion]: Information Locality as an Inductive Bias for Neural Language Models

  EkaterinaVoloshina started Feb 27, 2026 in Paper Suggestions

  

  

  0
• 2 You must be logged in to vote

  📆

  [Reading Group]: Randomly Removing 50% of Dimensions in Text Embeddings has Minimal Impact on Retrieval and Classification Tasks

  nicaudinet announced Jan 23, 2026 in Meetings & Announcements · Closed

  

  

  

  

  3
• 4 You must be logged in to vote

  📚

  [Paper Suggestion]: Randomly Removing 50% of Dimensions in Text Embeddings has Minimal Impact on Retrieval and Classification Tasks

  nicaudinet started Jan 7, 2026 in Paper Suggestions · Closed

  

  

  0
• 2 You must be logged in to vote

  📆

  [Reading Group]: RLP: Reinforcement as a Pretraining Objective

  ricj announced Dec 1, 2025 in Meetings & Announcements · Closed

  

  

  

  

  5
• 2 You must be logged in to vote

  📚

  [Paper Suggestion]: Nested Learning: The Illusion of Deep Learning Architectures

  EkaterinaVoloshina started Nov 27, 2025 in Paper Suggestions

  

  

  0
• 3 You must be logged in to vote

  📚

  [Paper Suggestion]: Weight-sparse transformers have interpretable circuits

  ricj started Nov 21, 2025 in Paper Suggestions · Closed

  

  

  0
• 3 You must be logged in to vote

  📚

  [Paper Suggestion]: Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model?

  ricj started Nov 21, 2025 in Paper Suggestions · Closed

  

  

  0
• 2 You must be logged in to vote

  📆

  [Reading Group]: Expanding the Action Space of LLMs to Reason Beyond Language

  ricj announced Nov 5, 2025 in Meetings & Announcements · Closed

  

  

  

  

  11
• 1 You must be logged in to vote

  📆

  [Meeting]: FlexOlmo: Open Language Models for Flexible Data Use

  ricj announced Oct 7, 2025 in Meetings & Announcements · Closed

  

  

  

  

  8
• 4 You must be logged in to vote

  📚

  [Paper Suggestion]: Large Language Model Hacking: Quantifying the Hidden Risks of Using LLMs for Text Annotation

  ricj started Oct 17, 2025 in Paper Suggestions

  

  

  0
• 2 You must be logged in to vote

  📚

  [Paper Suggestion]: Expanding the Action Space of LLMs to Reason Beyond Language

  ricj started Oct 17, 2025 in Paper Suggestions · Closed

  

  

  0
• 3 You must be logged in to vote

  📚

  [Paper Suggestion]: RLP: Reinforcement as a Pretraining Objective

  NicklasOsterbacka started Oct 6, 2025 in Paper Suggestions · Closed

  

  

  0
• 📚

  [Paper Suggestion]: FlexOlmo: Open Language Models for Flexible Data Use

  ricj started Oct 2, 2025 in Paper Suggestions · Closed

  

  

  0
• 1 You must be logged in to vote

  📆

  [Reading Group]: Learning Transformer Programs

  nicaudinet announced Sep 15, 2025 in Meetings & Announcements · Closed

  

  

  

  

  7
• 2 You must be logged in to vote

  📚

  [Paper Suggestion]: Why language models hallucinate

  m3hrdadfi started Sep 7, 2025 in Paper Suggestions · Closed

  

  

  0
• 3 You must be logged in to vote

  📚

  [Paper Suggestion]: Large Language Models Are Biased Because They Are Large Language Models

  nicaudinet started Sep 3, 2025 in Paper Suggestions · Closed

  

  

  0
• 1 You must be logged in to vote

  📚

  [Paper Suggestion]: Invite guest Shuai Wang

  ricj started Sep 1, 2025 in Paper Suggestions · Closed

  

  

  0
• 3 You must be logged in to vote

  📚

  [Paper Suggestion]: Learning Transformer Programs

  nicaudinet started Sep 1, 2025 in Paper Suggestions · Closed

  

  

  0
• 2 You must be logged in to vote

  📆

  [Reading Group]: Large Language Diffusion Models

  nicaudinet announced Aug 26, 2025 in Meetings & Announcements · Closed

  

  

  

  

  6
• 📚

  [Paper Suggestion]: Large Language Diffusion Models

  NicklasOsterbacka started Jul 16, 2025 in Paper Suggestions · Closed

  

  

  0