Best AI papers explained

471 Avsnitt

1. An empirical risk minimization approach for offline inverse RL and Dynamic Discrete Choice models

   Publicerades: 2025-07-22
2. Inverse Reinforcement Learning Meets Large Language Model Post-Training: Basics, Advances, and Opportunities

   Publicerades: 2025-07-22
3. The Invisible Leash: Why RLVR May Not Escape Its Origin

   Publicerades: 2025-07-20
4. Language Model Personalization via Reward Factorization

   Publicerades: 2025-07-20
5. Train for the Worst, Plan for the Best: Understanding Token Ordering in Masked Diffusions

   Publicerades: 2025-07-18
6. Do We Need to Verify Step by Step? Rethinking Process Supervision from a Theoretical Perspective

   Publicerades: 2025-07-17
7. Soft Best-of-n Sampling for Model Alignment

   Publicerades: 2025-07-16
8. On Temporal Credit Assignment and Data-Efficient Reinforcement Learning

   Publicerades: 2025-07-15
9. Bradley–Terry and Multi-Objective Reward Modeling Are Complementary

   Publicerades: 2025-07-15
10. Probing Foundation Models for World Models

    Publicerades: 2025-07-15
11. GenAI-Powered Statistical Inference (with Unstructured Data)

    Publicerades: 2025-07-14
12. Interpretable Reward Modeling with Active Concept Bottlenecks

    Publicerades: 2025-07-14
13. PrefillOnly: An Inference Engine for Prefill-only Workloads in Large Language Model Applications

    Publicerades: 2025-07-14
14. A Collectivist, Economic Perspective on AI

    Publicerades: 2025-07-14
15. Textual Bayes: Quantifying Uncertainty in LLM-Based Systems

    Publicerades: 2025-07-12
16. The Winner's Curse in Data-Driven Decisions

    Publicerades: 2025-07-11
17. SPIRAL: Self-Play for Reasoning Through Zero-Sum Games

    Publicerades: 2025-07-11
18. Beyond Statistical Learning: Exact Learning Is Essential for General Intelligence

    Publicerades: 2025-07-11
19. Aligning Learning and Endogenous Decision-Making

    Publicerades: 2025-07-11
20. Reliable Statistical Inference with Synthetic Data from Large Language Models

    Publicerades: 2025-07-11