Best AI papers explained

520 Afleveringen

1. Can Large reasoning models self-train?

   Gepubliceerd: 1-11-2025
2. ALITA-G: Self-Evolving Generative Agent for Agent Generation

   Gepubliceerd: 1-11-2025
3. Self-improving LLM agents at test-time

   Gepubliceerd: 30-10-2025
4. Offline RL by Reward-Weighted Fine-Tuning for Conversation Optimization

   Gepubliceerd: 30-10-2025
5. Language models are injective and hence invertible

   Gepubliceerd: 30-10-2025
6. ReasoningBank: Scaling Agent Self-Evolving with Reasoning Memory

   Gepubliceerd: 29-10-2025
7. RLAD: Training LLMs to Discover Abstractions

   Gepubliceerd: 29-10-2025
8. How to Train Your Advisor: Steering Black-Box LLMs with ADVISOR MODELS

   Gepubliceerd: 29-10-2025
9. Self-improving LLM agents at Test-Time

   Gepubliceerd: 27-10-2025
10. KL-Regularized Reinforcement Learning is designed to Mode Collapse

    Gepubliceerd: 27-10-2025
11. How do LLMs use their depth?

    Gepubliceerd: 27-10-2025
12. Thought Communication in Multiagent Collaboration

    Gepubliceerd: 27-10-2025
13. Reasoning with Sampling: Base Models Outperform RL

    Gepubliceerd: 26-10-2025
14. Continual Learning via Sparse Memory Finetuning

    Gepubliceerd: 26-10-2025
15. Direct Preference Optimization with Unobserved Preference Heterogeneity: The Necessity of Ternary Preferences

    Gepubliceerd: 24-10-2025
16. The Coverage Principle: How Pre-Training Enables Post-Training

    Gepubliceerd: 24-10-2025
17. The Era of Real-World Human Interaction: RL from User Conversations

    Gepubliceerd: 24-10-2025
18. Agent Learning via Early Experience

    Gepubliceerd: 24-10-2025
19. Demystifying the Mechanisms Behind Emergent Exploration in Goal-conditioned RL

    Gepubliceerd: 22-10-2025
20. Rewriting History: A Recipe for Interventional Analyses to Study Data Effects on Model Behavior

    Gepubliceerd: 22-10-2025