Med-Scout: 通过几何感知的强化学习后训练治愈多模态大语言模型在医学感知中的几何盲点
Med-Scout: Curing MLLMs' Geometric Blindness in Medical Perception via Geometry-Aware RL Post-Training
AI总结：
提出Med-Scout框架，利用无标注医学图像中的内在几何逻辑，通过强化学习和三种代理任务（层次尺度定位、拓扑拼图重建、异常一致性检测）来缓解多模态大语言模型的几何盲点，并在新基准Med-Scout-Bench上提升超过40%的几何感知性能，同时泛化到更广泛的医学理解任务。
摘要：
尽管最近的多模态大语言模型（MLLMs）在医学诊断中展现出语言能力，但我们发现即使是最先进的MLLMs也存在一个关键的感知缺陷：几何盲点。这种无法将输出基于客观几何约束的问题导致了看似合理但事实错误的幻觉，其根源在于训练范式优先考虑语言流畅性而非几何保真度。本文介绍了Med-Scout，一种新颖的框架，通过强化学习（RL）“治愈”这种盲点，利用未标记医学图像中内在的几何逻辑。Med-Scout不依赖昂贵的人工标注，而是通过受临床医生系统阅读和推理模式启发的三种策略性代理任务推导出可验证的监督信号：层次尺度定位、拓扑拼图重建和异常一致性检测。为了严格量化这一缺陷，我们提出了Med-Scout-Bench，一个专门设计用于评估几何感知的新基准。大量评估表明，Med-Scout显著缓解了几何盲点，在我们的基准上比领先的专有和开源MLLMs提升了超过40%。此外，这种增强的几何感知泛化到更广泛的医学理解，在放射学和综合性医学VQA任务上取得了优异结果。
英文摘要：
Despite recent Multimodal Large Language Models (MLLMs)' linguistic prowess in medical diagnosis, we find even state-of-the-art MLLMs suffer from a critical perceptual deficit: geometric blindness. This failure to ground outputs in objective geometric constraints leads to plausible yet factually incorrect hallucinations, rooted in training paradigms that prioritize linguistic fluency over geometric fidelity. This paper introduces Med-Scout, a novel framework that "cures" this blindness via Reinforcement Learning (RL) that leverages the intrinsic geometric logic latent within unlabeled medical images. Instead of relying on costly expert annotations, Med-Scout derives verifiable supervision signals through three strategic proxy tasks inspired by the systematic reading and reasoning patterns of clinicians: Hierarchical Scale Localization, Topological Jigsaw Reconstruction, and Anomaly Consistency Detection. To rigorously quantify this deficit, we present Med-Scout-Bench, a new benchmark specifically designed to evaluate geometric perception. Extensive evaluations show that Med-Scout significantly mitigates geometric blindness, outperforming leading proprietary and open-source MLLMs by over 40% on our benchmark. Furthermore, this enhanced geometric perception generalizes to broader medical understanding, achieving superior results on radiological and comprehensive medical VQA tasks.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
