Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning

Xie, Tian, Gao, Zitian, Ren, Qingnan · arXiv · 2025