Ximing Lu

Latest

• ProfBench: Multi-Domain Rubrics requiring Professional Knowledge to Answer and Judge
• ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models