515 Folgen

  1. Beyond Markovian: Reflective Exploration via Bayes-Adaptive RL for LLM Reasoning

    Vom: 29.5.2025
  2. Planning without Search: Refining Frontier LLMs with Offline Goal-Conditioned RL

    Vom: 29.5.2025
  3. Value-Guided Search for Efficient Chain-of-Thought Reasoning

    Vom: 29.5.2025
  4. Shallow Preference Signals: Large Language model aligns even better without truncated data?

    Vom: 29.5.2025
  5. Gaming Tool Preferences in Agentic LLMs

    Vom: 29.5.2025
  6. Partner Modelling Emerges in Recurrent Agents (But Only When It Matters)

    Vom: 29.5.2025
  7. LLM Populations Form Social Conventions and Collective Bias

    Vom: 29.5.2025
  8. LLM Generated Persona is a Promise with a Catch

    Vom: 29.5.2025
  9. Large Language Models for Digital Twin Simulation

    Vom: 29.5.2025
  10. From RL Distillation to Autonomous LLM Agents

    Vom: 29.5.2025
  11. Prompting, Auto-Prompting, and Human-AI Communication

    Vom: 29.5.2025
  12. Textual Gradients for LLM Optimization

    Vom: 29.5.2025
  13. Large Language Models as Markov Chains

    Vom: 28.5.2025
  14. Metastable Dynamics of Chain-of-Thought Reasoning: Provable Benefits of Search, RL and Distillation

    Vom: 28.5.2025
  15. Selective induction heads: how transformers select causal structures in context

    Vom: 28.5.2025
  16. The Evolution of Statistical Induction Heads: In-Context Learning Markov Chains

    Vom: 28.5.2025
  17. How Transformers Learn Causal Structure with Gradient Descent

    Vom: 28.5.2025
  18. Planning anything with rigor: general-purpose zero-shot planning with llm-based formalized programming

    Vom: 28.5.2025
  19. Automated Design of Agentic Systems

    Vom: 28.5.2025
  20. What’s the Magic Word? A Control Theory of LLM Prompting

    Vom: 28.5.2025

12 / 26

Cut through the noise. We curate and break down the most important AI papers so you don’t have to.

Visit the podcast's native language site