Best AI papers explained

456 Episodes

1. Deep Learning is Not So Mysterious or Different

   Published: 6/25/2025
2. AI Agents Need Authenticated Delegation

   Published: 6/25/2025
3. Probabilistic Modelling is Sufficient for Causal Inference

   Published: 6/25/2025
4. Not All Explanations for Deep Learning Phenomena Are Equally Valuable

   Published: 6/25/2025
5. e3: Learning to Explore Enables Extrapolation of Test-Time Compute for LLMs

   Published: 6/17/2025
6. Extrapolation by Association: Length Generalization Transfer in Transformers

   Published: 6/17/2025
7. Uncovering Causal Hierarchies in Language Model Capabilities

   Published: 6/17/2025
8. Generalization or Hallucination? Understanding Out-of-Context Reasoning in Transformers

   Published: 6/17/2025
9. Improving Treatment Effect Estimation with LLM-Based Data Augmentation

   Published: 6/17/2025
10. LLM Numerical Prediction Without Auto-Regression

    Published: 6/17/2025
11. Self-Adapting Language Models

    Published: 6/17/2025
12. Why in-context learning models are good few-shot learners?

    Published: 6/17/2025
13. Take Caution in Using LLMs as Human Surrogates: Scylla Ex Machina∗

    Published: 6/14/2025
14. The Logic of Machines: The AI Reasoning Debate

    Published: 6/12/2025
15. Layer by Layer: Uncovering Hidden Representations in Language Models

    Published: 6/12/2025
16. Causal Attribution Analysis for Continuous Outcomes

    Published: 6/12/2025
17. Training a Generally Curious Agent

    Published: 6/12/2025
18. Estimation of Treatment Effects Under Nonstationarity via Truncated Difference-in-Q’s

    Published: 6/12/2025
19. Strategy Coopetition Explains the Emergence and Transience of In-Context Learning

    Published: 6/12/2025
20. Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs

    Published: 6/11/2025