Jason Lee

7 papers Princeton University

• Accelerating RL for LLM Reasoning with Optimal Advantage Regression
• Deployment Efficient Reward-Free Exploration with Linear Function Approximation
• Emergence and scaling laws in SGD learning of shallow neural networks
• Learning Orthogonal Multi-Index Models: A Fine-Grained Information Exponent Analysis
• The Generative Leap: Tight Sample Complexity for Efficiently Learning Gaussian Multi-Index Models
• What Makes a Reward Model a Good Teacher? An Optimization Perspective
• What One Cannot, Two Can: Two-Layer Transformers Provably Represent Induction Heads on Any-Order Markov Chains