Best AI papers explained

522 Afleveringen

1. Value Flows: Flow-Based Distributional Reinforcement Learning

   Gepubliceerd: 14-10-2025
2. Self-Adapting Language Models

   Gepubliceerd: 12-10-2025
3. The Markovian Thinker

   Gepubliceerd: 12-10-2025
4. Moloch’s Bargain: emergent misalignment when LLMs compete for audiences

   Gepubliceerd: 12-10-2025
5. Transformer Predictor Dynamics and Task Diversity

   Gepubliceerd: 11-10-2025
6. Base models know how to reason, thinking models learn when

   Gepubliceerd: 11-10-2025
7. Spectrum tuning: Post-training for distributional coverage and in-context steerability

   Gepubliceerd: 11-10-2025
8. Understanding Prompt Tuning and In-Context Learning via Meta-Learning

   Gepubliceerd: 11-10-2025
9. MLPs Learn In-Context on Regression and Classification tasks

   Gepubliceerd: 11-10-2025
10. Is Pre-Training Truly Better than Meta-Learning?

    Gepubliceerd: 11-10-2025
11. Agentic Context Engineering: Evolving Contexts for Self-Improving Language Models

    Gepubliceerd: 11-10-2025
12. Do LLMs Recognize Your Preferences? Evaluating Personalized Preference Following in LLMs

    Gepubliceerd: 9-10-2025
13. Learning dynamics of LLM finetuning

    Gepubliceerd: 9-10-2025
14. Iterative Data Smoothing: Mitigating Reward Overfitting and Overoptimization in RLHF

    Gepubliceerd: 9-10-2025
15. OpenAI Agent Builder and n8n: Orchestrating Reasoning Versus Automating Process

    Gepubliceerd: 8-10-2025
16. Training Agents Inside of Scalable World Models

    Gepubliceerd: 8-10-2025
17. Small Language Models are the Future of Agentic AI

    Gepubliceerd: 7-10-2025
18. Activation Steering in Generative Settings via Contrastive Causal Mediation Analysis

    Gepubliceerd: 6-10-2025
19. Eliciting Secret Knowledge from Language Models

    Gepubliceerd: 6-10-2025
20. Temporal difference flow

    Gepubliceerd: 6-10-2025