arXiv：cs.AI· Karan Dua, Amit Agarwal, Hitesh Laxmichand Patel, Hansa Meghwani, Jyotika Singh, Ranjeet Gupta, Graham Horwood, Tao Sheng, Avi Sil, Sujith Ravi, Dan Roth·· 5 小时前
从人像到病理：探究视觉语言模型的诊断越界行为
From Portraits to Pathology: Probing Diagnostic Overreach in Vision-Language Models
AI 导读
研究构建了 PARITY 数据集，用于测试视觉语言模型（VLM）是否会对面部图像做出 ASD、ADHD、MDD 和 GAD 的诊断判断。对 27 个开源与闭源模型的评估显示，模型会将面部表情转化为缺乏依据的医学判断，这一行为在开源权重 VLM 中普遍存在，并延伸至经安全训练的前沿模型——它们在单张人像上拒绝作答，却在比较同一身份的多张人像时给出诊断。
正文
Authors:Karan Dua, Amit Agarwal, Hitesh Laxmichand Patel, Hansa Meghwani, Jyotika Singh, Ranjeet Gupta, Graham Horwood, Tao Sheng, Avi Sil, Sujith Ravi, Dan Roth
Abstract:Many medical conditions require clinical assessment of symptoms, history, and functioning rather than judgments from appearance. We study whether vision-language models (VLMs) abstain from facial attribution of autism spectrum disorder (ASD), attention-deficit/hyperactivity disorder (ADHD), major depressive disorder (MDD), and generalized anxiety disorder (GAD).
To support this study, we construct PARITY (Paired Assessment with Reused Identity), a demographically balanced dataset for testing whether changing facial expressions alters diagnostic judgments, while holding identity and other visual factors as constant as possible.
Our comprehensive evaluation of 27 open- and closed-weight models reveals inconsistent abstention across expressions, conditions, and task formats. Models turn facial expressions into unsupported medical judgments: this behavior is widespread among open-weight VLMs and extends to safety-trained frontier models that abstain on individual portraits yet make diagnostic judgments when comparing portraits of the same identity. Safeguard checks reveal further gaps: clinical instructions do not consistently prevent these judgments, and a guardrail model rarely flags paired requests as unsafe. Semantic analysis further shows that most baseline open-model refusals fail to explain both the insufficiency of facial evidence and the need for clinical assessment. Unsupported judgments and seemingly authoritative explanations can cause distress, reinforce stigma, and offer false reassurance, underscoring the need for stronger safety measures in model training and deployment.
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| ACM classes: | I.2.7; I.2.10 |
| Cite as: | arXiv:2609.05540 [cs.CV] |
| (or arXiv:2609.05540v2 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2609.05540 arXiv-issued DOI via DataCite |
Submission history
From: Karan Dua [view email]
[v1]
Wed, 2 Sep 2026 11:52:02 UTC (8,226 KB)
[v2]
Mon, 12 Oct 2026 10:50:51 UTC (1,543 KB)
来源：arXiv：cs.AI · arxiv.org