Wenhao Zhan

Accelerating RL for LLM Reasoning with Optimal Advantage Regression

Exploiting Structure in Offline Multi-Agent RL: The Benefits of Low Interaction Rank

Regressing the Relative Future: Efficient Policy Optimization for Multi-turn RLHF

Correcting the Mythos of KL-Regularization: Direct Alignment without Overoptimization via Chi-Squared Preference Optimization

Provable Offline Preference-Based Reinforcement Learning

Provable Reward-Agnostic Preference-Based Reinforcement Learning

REBEL: Reinforcement Learning via Regressing Relative Rewards

Provably Efficient CVaR RL in Low-rank MDPs