Archives

The club concluded in August 2026 after 22 sessions. See the closing announcement.

2026

• LLMs Get Lost in Multi-Turn Conversation
• Why Do Multi-Agent LLM Systems Fail?
• SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training

2025

• Efficient GPT-4V Level Multimodal Large Language Model for Deployment on Edge Devices
• The Illusion of Thinking: Understanding the Strengths and Limitations of Reasoning Models via the Lens of Problem Complexity
• AI as Normal Technology
• On the Theoretical Limitations of Embedding-Based Retrieval

2024

• Modeling Tabular Data using Conditional GAN
• Gorilla: Large Language Model Connected with Massive APIs
• Dissecting Racial Bias in an Algorithm Used to Manage the Health of Populations
• Are Emergent Abilities of Large Language Models a Mirage?
• Direct Preference Optimization: Your Language Model is Secretly a Reward Model

2023

• Why Do Tree-based Models Still Outperform Deep Learning on Typical Tabular Data?
• Neural Machine Translation by Jointly Learning to Align and Translate
• Constitutional AI: Harmlessness from AI Feedback
• AutoKnow: Self-Driving Knowledge Collection for Products of Thousands of Types
• Let’s Verify Step by Step
• Visualization in Bayesian workflow
• Reviewing a Case Study About Real Estate Market Prediction
• Leakage and the Reproducibility Crisis in ML-based Science

2022

• Zero-Shot Text-to-Image Generation
• “Everyone wants to do the model work, not the data work”: Data Cascades in High-Stakes AI