Best AI papers explained

550 Episodes

1. Uncovering Causal Hierarchies in Language Model Capabilities

   Published: 17/06/2025
2. Generalization or Hallucination? Understanding Out-of-Context Reasoning in Transformers

   Published: 17/06/2025
3. Improving Treatment Effect Estimation with LLM-Based Data Augmentation

   Published: 17/06/2025
4. LLM Numerical Prediction Without Auto-Regression

   Published: 17/06/2025
5. Self-Adapting Language Models

   Published: 17/06/2025
6. Why in-context learning models are good few-shot learners?

   Published: 17/06/2025
7. Take Caution in Using LLMs as Human Surrogates: Scylla Ex Machina∗

   Published: 14/06/2025
8. The Logic of Machines: The AI Reasoning Debate

   Published: 12/06/2025
9. Layer by Layer: Uncovering Hidden Representations in Language Models

   Published: 12/06/2025
10. Causal Attribution Analysis for Continuous Outcomes

    Published: 12/06/2025
11. Training a Generally Curious Agent

    Published: 12/06/2025
12. Estimation of Treatment Effects Under Nonstationarity via Truncated Difference-in-Q’s

    Published: 12/06/2025
13. Strategy Coopetition Explains the Emergence and Transience of In-Context Learning

    Published: 12/06/2025
14. Emergent Misalignment: Narrow finetuning can produce broadly misaligned LLMs

    Published: 11/06/2025
15. Agentic Supernet for Multi-agent Architecture Search

    Published: 11/06/2025
16. Sample Complexity and Representation Ability of Test-time Scaling Paradigms

    Published: 11/06/2025
17. Aligning with Human Judgement: The Role of Pairwise Preference in Large Language Model Evaluators

    Published: 10/06/2025
18. LLMs Get Lost In Multi-Turn Conversation

    Published: 09/06/2025
19. PromptPex: Automatic Test Generation for Prompts

    Published: 08/06/2025
20. General Agents Need World Models

    Published: 08/06/2025