Best AI papers explained

529 Jaksot

1. Markets with Heterogeneous Agents: Dynamics and Survival of Bayesian vs. No-Regret Learners

   Julkaistiin: 5.7.2025
2. Why Neural Network Can Discover Symbolic Structures with Gradient-based Training: An Algebraic and Geometric Foundation

   Julkaistiin: 5.7.2025
3. Causal Abstraction with Lossy Representations

   Julkaistiin: 4.7.2025
4. The Winner's Curse in Data-Driven Decisions

   Julkaistiin: 4.7.2025
5. Embodied AI Agents: Modeling the World

   Julkaistiin: 4.7.2025
6. Beyond Statistical Learning: Exact Learning Is Essential for General Intelligence

   Julkaistiin: 4.7.2025
7. What Has a Foundation Model Found? Inductive Bias Reveals World Models

   Julkaistiin: 4.7.2025
8. Language Bottleneck Models: A Framework for Interpretable Knowledge Tracing and Beyond

   Julkaistiin: 3.7.2025
9. Learning to Explore: An In-Context Learning Approach for Pure Exploration

   Julkaistiin: 3.7.2025
10. Human-AI Matching: The Limits of Algorithmic Search

    Julkaistiin: 25.6.2025
11. Uncertainty Quantification Needs Reassessment for Large-language Model Agents

    Julkaistiin: 25.6.2025
12. Bayesian Meta-Reasoning for Robust LLM Generalization

    Julkaistiin: 25.6.2025
13. General Intelligence Requires Reward-based Pretraining

    Julkaistiin: 25.6.2025
14. Deep Learning is Not So Mysterious or Different

    Julkaistiin: 25.6.2025
15. AI Agents Need Authenticated Delegation

    Julkaistiin: 25.6.2025
16. Probabilistic Modelling is Sufficient for Causal Inference

    Julkaistiin: 25.6.2025
17. Not All Explanations for Deep Learning Phenomena Are Equally Valuable

    Julkaistiin: 25.6.2025
18. e3: Learning to Explore Enables Extrapolation of Test-Time Compute for LLMs

    Julkaistiin: 17.6.2025
19. Extrapolation by Association: Length Generalization Transfer in Transformers

    Julkaistiin: 17.6.2025
20. Uncovering Causal Hierarchies in Language Model Capabilities

    Julkaistiin: 17.6.2025