ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models

AI Breakdown - ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models