Fetching the paper…

Point-RFT: Improving Multimodal Reasoning with Visually Grounded Reinforcement Finetuning · Around