Lipeng Xie

Auto-Rubric: Learning to Extract Generalizable Criteria for Reward Modeling

Taming the Judge: Deconflicting AI Feedback for Stable Reinforcement Learning