Best AI papers explained

454 Episodes

1. Actor-Critic without Actor: Critic-Guided Denoising for RL

   Published: 9/29/2025
2. DELTA-Code: How Does RL Unlock and Transfer New Programming Algorithms in LLMs?

   Published: 9/29/2025
3. Linear Transformers Implicitly Discover Unified Numerical Algorithms

   Published: 9/29/2025
4. Regularizing Extrapolation in Causal Inference

   Published: 9/27/2025
5. DoubleGen - Debiased Generative Modeling of Counterfactuals

   Published: 9/27/2025
6. What Characterizes Effective Reasoning? Revisiting Length, Review, and Structure of CoT

   Published: 9/27/2025
7. Compute as Teacher: Turning Inference Compute Into Reference-Free Supervision

   Published: 9/27/2025
8. Learning without training: The implicit dynamics of in-context learning

   Published: 9/24/2025
9. Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model

   Published: 9/24/2025
10. Open Problems in Mechanistic Interpretability

    Published: 9/21/2025
11. Maestro: Joint Graph & Config Optimization for Reliable AI Agents

    Published: 9/21/2025
12. Thought Anchors: Which LLM Reasoning Steps Matter?

    Published: 9/21/2025
13. Sample Complexity and Representation Ability of Test-time Scaling Paradigms

    Published: 9/9/2025
14. RL's Razor: Why Online RL Forgets Less

    Published: 9/7/2025
15. Why Language Models Hallucinate

    Published: 9/6/2025
16. ALFA: Aligning LLMs to Ask Good Questions A Case Study in Clinical Reasoning

    Published: 9/6/2025
17. Sample Efficient Preference Alignment in LLMs via Active Exploration

    Published: 9/6/2025
18. Adventures in Demand Analysis Using AI

    Published: 9/4/2025
19. Memento: Fine-tuning LLM Agents without Fine-tuning LLMs

    Published: 9/1/2025
20. On the Theoretical Limitations of Embedding-Based Retrieval

    Published: 8/31/2025