Fetching the paper…

R1-Code-Interpreter: LLMs Reason with Code via Supervised and Multi-stage Reinforcement Learning · Around