Fetching the paper…

OThink-MR1: Stimulating multimodal generalized reasoning capabilities via dynamic reinforcement learning · Around