Best AI papers explained

508 Episodo

1. Compute as Teacher: Turning Inference Compute Into Reference-Free Supervision

   Publicado: 27/9/2025
2. Learning without training: The implicit dynamics of in-context learning

   Publicado: 24/9/2025
3. Does Reinforcement Learning Really Incentivize Reasoning Capacity in LLMs Beyond the Base Model

   Publicado: 24/9/2025
4. Open Problems in Mechanistic Interpretability

   Publicado: 21/9/2025
5. Maestro: Joint Graph & Config Optimization for Reliable AI Agents

   Publicado: 21/9/2025
6. Thought Anchors: Which LLM Reasoning Steps Matter?

   Publicado: 21/9/2025
7. Sample Complexity and Representation Ability of Test-time Scaling Paradigms

   Publicado: 9/9/2025
8. RL's Razor: Why Online RL Forgets Less

   Publicado: 7/9/2025
9. Why Language Models Hallucinate

   Publicado: 6/9/2025
10. ALFA: Aligning LLMs to Ask Good Questions A Case Study in Clinical Reasoning

    Publicado: 6/9/2025
11. Sample Efficient Preference Alignment in LLMs via Active Exploration

    Publicado: 6/9/2025
12. Adventures in Demand Analysis Using AI

    Publicado: 4/9/2025
13. Memento: Fine-tuning LLM Agents without Fine-tuning LLMs

    Publicado: 1/9/2025
14. On the Theoretical Limitations of Embedding-Based Retrieval

    Publicado: 31/8/2025
15. Performance Prediction for Large Systems via Text-to-Text Regression

    Publicado: 30/8/2025
16. Demystifying the Visual Quality Paradox in Multimodal Large Language Models

    Publicado: 30/8/2025
17. Chain-of-Agents: End-to-End Agent Foundation Models via Multi-Agent Distillation and Agentic RL

    Publicado: 30/8/2025
18. Compute-Optimal Scaling for Value-Based Deep RL

    Publicado: 25/8/2025
19. LLM-based Conversational Recommendation Agents with Collaborative Verbalized Experience

    Publicado: 23/8/2025
20. Signal and Noise: Evaluating Language Model Benchmarks

    Publicado: 23/8/2025