让医疗视觉语言模型更依赖真实病灶证据，避免靠套路答题。
Med-OPD: Improving Medical Vision-Language Models via Evidence-Aware On-Policy Distillation
- 用证据感知信号引导训练，聚焦关键病灶区域
- 在多个医学影像任务上超越监督微调和普通蒸馏方法
- 适合希望提升AI诊断可信度的研究者和临床开发者
医疗视觉语言模型（Med-VLMs）需基于细粒度视觉证据进行可靠推理，但现有模型常依赖语言先验或医学模板生成看似合理却非基于真实病灶的结论。在策略蒸馏（OPD）中，学生模型生成的轨迹获得密集的标记级监督，且无需共享原始患者数据，具备隐私保护优势。然而，传统OPD对所有令牌均匀蒸馏，导致稀疏的证据依赖性令牌被大量临床描述性令牌稀释。受大语言模型中OPD成功的启发，我们提出首个统一的后训练框架——Med-OPD，将策略蒸馏与医疗证据感知监督结合。引入医学证据优势（MEA），一种基于教师模型的反事实信号，通过答案感知提示聚焦于支持目标诊断的证据，并通过比较原始图像与证据退化图像下的教师似然值，衡量每个令牌对医学视觉证据的依赖程度。基于MEA，Med-OPD在令牌和轨迹层面重新分配蒸馏信号，强化诊断关键令牌与证据依赖性推理路径。在OmniMedVQA子集上的实验表明，Med-OPD在CT、MRI、疾病诊断和病灶分级任务上均持续优于SFT和标准OPD。结果证明，证据感知蒸馏能更好增强模型对关键视觉证据的依赖，提升多模态医疗推理的可靠性。代码与数据已公开：https://github.com/yunhang8658/MedOPD.git
原文摘要 · Abstract (English)
Medical Vision-Language Models (Med-VLMs) require reliable reasoning from fine-grained visual evidence, yet existing models can produce plausible clinical answers by relying on language priors or medical templates rather than truly attending to diagnosis-critical regions. On-Policy Distillation (OPD) offers dense token-level supervision on student-generated trajectories and provides a privacy-compatible means of capability transfer without requiring the redistribution of raw patient data. However, standard OPD uniformly distills all tokens, causing sparse evidence-dependent tokens to be diluted by abundant clinical narrative tokens. Inspired by the success of OPD in the large language model community, we propose \textbf{Med-OPD}, to our knowledge the first unified post-training framework that integrates on-policy distillation with medical evidence-aware supervision for Med-VLMs. We introduce \textbf{Medical Evidence Advantage} (MEA), a teacher-grounded counterfactual signal that uses an answer-aware hint to focus teacher scoring on evidence supporting the target diagnosis, and measures each token's dependence on medical visual evidence by comparing teacher likelihoods under the original and evidence-degraded imaging modalities. Based on MEA, Med-OPD redistributes the distillation signal at both the token and trajectory levels, emphasizing diagnosis-critical tokens and evidence-reliant rollouts. Experiments on OmniMedVQA subsets show that Med-OPD consistently outperforms SFT and standard OPD across CT, MRI, Disease Diagnosis, and Lesion Grading. These results demonstrate that evidence-aware distillation can better strengthen medical VLMs' reliance on key visual evidence and improve reliable multimodal medical reasoning. The source code and data is publicly available at: https://github.com/yunhang8658/MedOPD.git
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。