R1-ShareVL: Incentivizing Reasoning Capability of Multimodal Large Language Models via Share-GRPO Paper • 2505.16673 • Published May 22 • 2