MM-PRM: Enhancing Multimodal Mathematical Reasoning with Scalable Step-Level Supervision Paper • 2505.13427 • Published May 19 • 26
CPGD: Toward Stable Rule-based Reinforcement Learning for Language Models Paper • 2505.12504 • Published May 18 • 24
MM-Eureka: Exploring Visual Aha Moment with Rule-based Large-scale Reinforcement Learning Paper • 2503.07365 • Published Mar 10 • 62