Fetching the paper…

R1-Reward: Training Multimodal Reward Model Through Stable Reinforcement Learning · Around