Jinyi Han

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

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

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

HINT: Helping Ineffective Rollouts Navigate Towards Effectiveness

Think Thrice Before You Act: Progressive Thought Refinement in Large Language Models