Sihang Jiang

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

Higher-Order Cognitive Chain-of-Thought Is Enough: Evaluation, Analysis, and 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

HINT: Helping Ineffective Rollouts Navigate Towards Effectiveness

RL Beats SFT while Mitigating Definition Bias in LLM-based Information Extraction

Structure-Rich Text Benchmark for Knowledge Inference Evaluation