Best AI papers explained

526 Episodes

1. Survey of In-Context Learning Interpretation and Analysis

   Published: 23/05/2025
2. From Decoding to Meta-Generation: Inference-time Algorithms for Large Language Models

   Published: 23/05/2025
3. LLM In-Context Learning as Kernel Regression

   Published: 23/05/2025
4. Where does In-context Learning Happen in Large Language Models?

   Published: 23/05/2025
5. Auto-Differentiating Any LLM Workflow: A Farewell to Manual Prompting

   Published: 22/05/2025
6. metaTextGrad: Learning to learn with language models as optimizers

   Published: 22/05/2025
7. Semantic Operators: A Declarative Model for Rich, AI-based Data Processing

   Published: 22/05/2025
8. Isolated Causal Effects of Language

   Published: 22/05/2025
9. Sleep-time Compute: Beyond Inference Scaling at Test-time

   Published: 22/05/2025
10. J1: Incentivizing Thinking in LLM-as-a-Judge

    Published: 22/05/2025
11. ShiQ: Bringing back Bellman to LLMs

    Published: 22/05/2025
12. Policy Learning with a Natural Language Action Space: A Causal Approach

    Published: 22/05/2025
13. Multi-Objective Preference Optimization: Improving Human Alignment of Generative Models

    Published: 22/05/2025
14. End-to-End Learning for Stochastic Optimization: A Bayesian Perspective

    Published: 21/05/2025
15. TEXTGRAD: Automatic Differentiation via Text

    Published: 21/05/2025
16. Steering off Course: Reliability Challenges in Steering Language Models

    Published: 20/05/2025
17. Past-Token Prediction for Long-Context Robot Policies

    Published: 20/05/2025
18. Recovering Coherent Event Probabilities from LLM Embeddings

    Published: 20/05/2025
19. Systematic Meta-Abilities Alignment in Large Reasoning Models

    Published: 20/05/2025
20. Predictability Shapes Adaptation: An Evolutionary Perspective on Modes of Learning in Transformers

    Published: 20/05/2025