Best AI papers explained

441 Episodo

1. Emergent Strategic AI Equilibrium from Pre-trained Reasoning

   Publicado: 7/5/2025
2. Benefiting from Proprietary Data with Siloed Training

   Publicado: 6/5/2025
3. Advantage Alignment Algorithms

   Publicado: 6/5/2025
4. Asymptotic Safety Guarantees Based On Scalable Oversight

   Publicado: 6/5/2025
5. What Makes a Reward Model a Good Teacher? An Optimization Perspective

   Publicado: 6/5/2025
6. Towards Guaranteed Safe AI: A Framework for Ensuring Robust and Reliable AI Systems

   Publicado: 6/5/2025
7. Identifiable Steering via Sparse Autoencoding of Multi-Concept Shifts

   Publicado: 6/5/2025
8. You Are What You Eat - AI Alignment Requires Understanding How Data Shapes Structure and Generalisation

   Publicado: 6/5/2025
9. Interplay of LLMs in Information Retrieval Evaluation

   Publicado: 3/5/2025
10. Trade-Offs Between Tasks Induced by Capacity Constraints Bound the Scope of Intelligence

    Publicado: 3/5/2025
11. Toward Efficient Exploration by Large Language Model Agents

    Publicado: 3/5/2025
12. Getting More Juice Out of the SFT Data: Reward Learning from Human Demonstration Improves SFT

    Publicado: 2/5/2025
13. Self-Consuming Generative Models with Curated Data

    Publicado: 2/5/2025
14. Bootstrapping Language Models with DPO Implicit Rewards

    Publicado: 2/5/2025
15. DeepSeek-Prover-V2: Advancing Formal Reasoning

    Publicado: 1/5/2025
16. THINKPRM: Data-Efficient Process Reward Models

    Publicado: 1/5/2025
17. Societal Frameworks and LLM Alignment

    Publicado: 29/4/2025
18. Risks from Multi-Agent Advanced AI

    Publicado: 29/4/2025
19. Causality-Aware Alignment for Large Language Model Debiasing

    Publicado: 29/4/2025
20. Reward Models Evaluate Consistency, Not Causality

    Publicado: 28/4/2025