多模态多智能体强化学习用于放射学报告生成：类似放射科医师的工作流程与临床可验证奖励

A novel multi-modal multi-agent reinforcement learning framework, MARL-Rad, is proposed for radiology report generation. This framework coordinates region-specific agents and a global integrating agent, optimized through clinically verifiable rewards. Unlike previous approaches that rely on single-model reinforcement learning or post-hoc agentization of independently trained models, this method jointly trains multiple agents and optimizes the entire agent system through reinforcement learning. Experiments on the MIMIC-CXR and IU X-ray datasets demonstrate that MARL-Rad consistently improves clinical efficacy (CE) metrics, such as RadGraph, CheXbert, and GREEN scores, achieving state-of-the-art CE performance. Further analyses confirm that MARL-Rad enhances laterality consistency and produces higher quality reports.