Light

Dark

Past Reading Groups By Topic

LLMs

Feb 20, 2024 Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
Feb 6, 2024 Causal parrots: Large language models may talk causality but are not causal

NLP

Mar 27, 2024 Underspecification Presents Challenges for Credibility in Modern Machine Learning
Feb 20, 2024 Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training
Feb 6, 2024 Causal parrots: Large language models may talk causality but are not causal

adversarial

Apr 17, 2024 Risks From Learned Optimization in Advanced Machine Learning Systems
Feb 20, 2024 Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

causality

Apr 3, 2024 Causal Fairness Field Guide: Perspectives from Social and Formal Sciences
Feb 6, 2024 Causal parrots: Large language models may talk causality but are not causal

computer_vision

Mar 27, 2024 Underspecification Presents Challenges for Credibility in Modern Machine Learning

ethics

Oct 8, 2024 AI Art is Theft: Labour, Extraction, and Exploitation Or, On the Dangers of Stochastic Pollocks
Jun 26, 2024 AI Art and its Impact on Artists
Apr 3, 2024 Causal Fairness Field Guide: Perspectives from Social and Formal Sciences

fairness

Jun 26, 2024 AI Art and its Impact on Artists
Apr 3, 2024 Causal Fairness Field Guide: Perspectives from Social and Formal Sciences
Mar 27, 2024 Underspecification Presents Challenges for Credibility in Modern Machine Learning
Mar 13, 2024 Arbitrariness and Social Prediction: The Confounding Role of Variance in Fair Classification
Jan 30, 2024 Differentially Private Fair Learning

interpretability

Oct 1, 2024 Towards A Rigorous Science of Interpretable Machine Learning
Apr 24, 2024 Probabilistic Dataset Reconstruction from Interpretable Models
Feb 13, 2024 Model Explanations with Differential Privacy

non-technical

Oct 1, 2024 Towards A Rigorous Science of Interpretable Machine Learning
Mar 5, 2024 Runaround + Code 8
Feb 27, 2024 Runaround - Isaac Asimov
Jan 24, 2024 Welcome / Intro

philosophy

Mar 27, 2024 Underspecification Presents Challenges for Credibility in Modern Machine Learning
Mar 5, 2024 Runaround + Code 8
Feb 27, 2024 Runaround - Isaac Asimov

privacy

Sep 24, 2024 Machine Unlearning
Jun 26, 2024 AI Art and its Impact on Artists
May 1, 2024 Evaluating the Impact of Local Differential Privacy on Utility Loss via Influence Functions
Apr 24, 2024 Probabilistic Dataset Reconstruction from Interpretable Models
Feb 13, 2024 Model Explanations with Differential Privacy
Jan 30, 2024 Differentially Private Fair Learning

safety

Apr 17, 2024 Risks From Learned Optimization in Advanced Machine Learning Systems
Feb 20, 2024 Sleeper Agents: Training Deceptive LLMs that Persist Through Safety Training

underspecification

Mar 27, 2024 Underspecification Presents Challenges for Credibility in Modern Machine Learning