Best AI papers explained

550 Episodes

1. Partner Modelling Emerges in Recurrent Agents (But Only When It Matters)

   Published: 5/29/2025
2. LLM Populations Form Social Conventions and Collective Bias

   Published: 5/29/2025
3. LLM Generated Persona is a Promise with a Catch

   Published: 5/29/2025
4. Large Language Models for Digital Twin Simulation

   Published: 5/29/2025
5. From RL Distillation to Autonomous LLM Agents

   Published: 5/29/2025
6. Prompting, Auto-Prompting, and Human-AI Communication

   Published: 5/29/2025
7. Textual Gradients for LLM Optimization

   Published: 5/29/2025
8. Large Language Models as Markov Chains

   Published: 5/28/2025
9. Metastable Dynamics of Chain-of-Thought Reasoning: Provable Benefits of Search, RL and Distillation

   Published: 5/28/2025
10. Selective induction heads: how transformers select causal structures in context

    Published: 5/28/2025
11. The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains

    Published: 5/28/2025
12. How Transformers Learn Causal Structure with Gradient Descent

    Published: 5/28/2025
13. Planning anything with rigor: general-purpose zero-shot planning with llm-based formalized programming

    Published: 5/28/2025
14. Automated Design of Agentic Systems

    Published: 5/28/2025
15. What’s the Magic Word? A Control Theory of LLM Prompting

    Published: 5/28/2025
16. BoNBoN Alignment for Large Language Models and the Sweetness of Best-of-n Sampling

    Published: 5/27/2025
17. RL with KL penalties is better viewed as Bayesian inference

    Published: 5/27/2025
18. Asymptotics of Language Model Alignment

    Published: 5/27/2025
19. Qwen 2.5, RL, and Random Rewards

    Published: 5/27/2025
20. Theoretical guarantees on the best-of-n alignment policy

    Published: 5/27/2025