Jiaqing Liang

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

OS-MAP: How Far Can Computer-Using Agents Go in Breadth and Depth?

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

Text-to-SQL Benchmarks for Enterprise Realities: Under Massive Scopes, Complex Schemas and Scattered Knowledge

Learning to Think in Blocks: A Prior-Guided Reinforcement Learning Framework for RAG

HINT: Helping Ineffective Rollouts Navigate Towards Effectiveness

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

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