Best AI papers explained

478 Episodes

1. Transformer Predictor Dynamics and Task Diversity

   Published: 10/11/2025
2. Base models know how to reason, thinking models learn when

   Published: 10/11/2025
3. Spectrum tuning: Post-training for distributional coverage and in-context steerability

   Published: 10/11/2025
4. Understanding Prompt Tuning and In-Context Learning via Meta-Learning

   Published: 10/11/2025
5. MLPs Learn In-Context on Regression and Classification tasks

   Published: 10/11/2025
6. Is Pre-Training Truly Better than Meta-Learning?

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

   Published: 10/11/2025
8. Do LLMs Recognize Your Preferences? Evaluating Personalized Preference Following in LLMs

   Published: 10/9/2025
9. Learning dynamics of LLM finetuning

   Published: 10/9/2025
10. Iterative Data Smoothing: Mitigating Reward Overfitting and Overoptimization in RLHF

    Published: 10/9/2025
11. OpenAI Agent Builder and n8n: Orchestrating Reasoning Versus Automating Process

    Published: 10/8/2025
12. Training Agents Inside of Scalable World Models

    Published: 10/8/2025
13. Small Language Models are the Future of Agentic AI

    Published: 10/7/2025
14. Activation Steering in Generative Settings via Contrastive Causal Mediation Analysis

    Published: 10/6/2025
15. Eliciting Secret Knowledge from Language Models

    Published: 10/6/2025
16. Temporal difference flow

    Published: 10/6/2025
17. Personalized reasoning: just-in-time personalization and why LLMs fail at it

    Published: 10/5/2025
18. Prompt Curriculum Learning for Efficient LLM Post-Training

    Published: 10/5/2025
19. Personalizing Reinforcement Learning from Human Feedback with Variational Preference Learning

    Published: 10/4/2025
20. Enhancing Personalized Multi-Turn Dialogue with Curiosity Reward

    Published: 10/4/2025