SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution Paper • 2502.18449 • Published Feb 25 • 75
