Siliang Zeng

Aligning Frozen LLMs by Reinforcement Learning: An Iterative Reweight-then-Optimize Approach

Reinforcing Multi-Turn Reasoning in LLM Agents via Turn-Level Reward Design

Joint Reward and Policy Learning with Demonstrations and Human Feedback Improves Alignment

From Demonstrations to Rewards: Alignment Without Explicit Human Preferences

Policy optimization can be memory-efficient: LLM Alignment Through Successive Policy Re-weighting (SPR)