Skywork R1V2: Multimodal Hybrid Reinforcement Learning for Reasoning Paper • 2504.16656 • Published Apr 23 • 56