GoT-R1: Unleashing Reasoning Capability of MLLM for Visual Generation with Reinforcement Learning Paper • 2505.17022 • Published May 22 • 27 • 2