Best AI papers explained

529 Episodes

1. Data Mixture Optimization: A Multi-fidelity Multi-scale Bayesian Framework

   Published: 31/03/2025
2. Why MCP won

   Published: 31/03/2025
3. SWEET-RL: Training LLM Agents for Collaborative Reasoning

   Published: 31/03/2025
4. TheoryCoder: Bilevel Planning with Synthesized World Models

   Published: 30/03/2025
5. Driving Forces in AI: Scaling to 2025 and Beyond (Jason Wei, OpenAI)

   Published: 29/03/2025
6. Expert Demonstrations for Sequential Decision Making under Heterogeneity

   Published: 28/03/2025
7. TextGrad: Backpropagating Language Model Feedback for Generative AI Optimization

   Published: 27/03/2025
8. MemReasoner: Generalizing Language Models on Reasoning-in-a-Haystack Tasks

   Published: 27/03/2025
9. RAFT: In-Domain Retrieval-Augmented Fine-Tuning for Language Models

   Published: 27/03/2025
10. Inductive Biases for Exchangeable Sequence Modeling

    Published: 26/03/2025
11. InverseRLignment: LLM Alignment via Inverse Reinforcement Learning

    Published: 26/03/2025
12. Prompt-OIRL: Offline Inverse RL for Query-Dependent Prompting

    Published: 26/03/2025
13. Alignment from Demonstrations for Large Language Models

    Published: 25/03/2025
14. Q♯: Distributional RL for Optimal LLM Post-Training

    Published: 18/03/2025
15. Scaling Test-Time Compute Without Verification or RL is Suboptimal

    Published: 14/03/2025
16. Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuning

    Published: 14/03/2025
17. Optimizing Test-Time Compute via Meta Reinforcement Fine-Tuning

    Published: 14/03/2025
18. Open Problems and Fundamental Limitations of Reinforcement Learning from Human Feedback

    Published: 14/03/2025
19. Revisiting Superficial Alignment Hypothesis

    Published: 14/03/2025
20. Diagnostic uncertainty: teaching language Models to describe open-ended uncertainty

    Published: 14/03/2025