Fei Yu

A Stitch in Time Saves Nine: Proactive Self-Refinement for Language Models

MAPO: Momentum-Aware Policy Optimization

Selective Expert Guidance for Effective and Diverse Exploration in Reinforcement Learning of LLMs

Your Models Have Thought Enough: Training Large Reasoning Models to Stop Overthinking

Automatic Robustness Stress Testing of LLMs as Mathematical Problem Solvers

HINT: Helping Ineffective Rollouts Navigate Towards Effectiveness

A Data-Centric Approach for Financial Large Language Models with Abductive Augmentation Reasoning