Best AI papers explained

534 Épisodes

1. Understanding neural networks through sparse circuits

   Publié: 14/11/2025
2. Supervised Reinforcement Learning: From Expert Trajectories to Step-wise Reasoning

   Publié: 14/11/2025
3. Multi-Agent Evolve: LLM Self-Improvement Through Co-Evolution

   Publié: 14/11/2025
4. LeJEPA: Provable and Scalable Self-Supervised Learning Without the Heuristics

   Publié: 14/11/2025
5. PREFDISCO: Evaluating Proactive Personalization through Interactive Preference Discovery

   Publié: 12/11/2025
6. Reusing pre-training data at test time is a compute multiplier

   Publié: 10/11/2025
7. Scaling Agent Learning via Experience Synthesis

   Publié: 09/11/2025
8. Continuous Autoregressive Language Models

   Publié: 08/11/2025
9. Toward a Theory of Agents as Tool-Use Decision-Makers

   Publié: 07/11/2025
10. Nested Learning: The Illusion of Deep Learning Architectures

    Publié: 05/11/2025
11. GST-UNet: A Neural Framework for Spatiotemporal Causal Inference with Time-Varying Confounding

    Publié: 05/11/2025
12. Beyond a million tokens: benchmarking and enhancing long-term memory in llms

    Publié: 04/11/2025
13. Agentic Economic Modeling

    Publié: 03/11/2025
14. Emergent Introspective Awareness in Large Language Models

    Publié: 03/11/2025
15. Can Large reasoning models self-train?

    Publié: 01/11/2025
16. ALITA-G: Self-Evolving Generative Agent for Agent Generation

    Publié: 01/11/2025
17. Self-improving LLM agents at test-time

    Publié: 30/10/2025
18. Offline RL by Reward-Weighted Fine-Tuning for Conversation Optimization

    Publié: 30/10/2025
19. Language models are injective and hence invertible

    Publié: 30/10/2025
20. ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory

    Publié: 29/10/2025