arXiv cs.CY· Soumyajit Datta, Arka Dutta, Ashiqur R. KhudaBukhsh·· 11 小时前AI 评分42
视觉语言模型如何通过图像变换编码种族与性取向等敏感属性
Physiognomy Revisited: How Vision-Language Models Encode Sensitive Attributes Through Image Transformation
AI 导读
研究发现两个知名视觉语言模型在被要求将人脸图像转换为特定种族（如西班牙裔、黑人、白人、亚裔）或性取向（如同性恋、异性恋）时，往往会照做，且是否遵从请求具有非确定性、群体敏感性，生成图像在统计上可区分并常强化刻板印象。图像生成中引入的表征偏差还会在文本推理与人物刻画任务中重现。这一现象不限于种族与性取向，图像生成同样会延续与犯罪相关的偏见，带来 AI 辅助画像的风险。
正文
Abstract:Emergent generative AI technologies, while remarkably beneficial across several tasks and domains, can suffer from unforeseen vulnerabilities and difficult-to-anticipate failure modes. Such failure modes can disparately affect vulnerable communities. This paper investigates two well-known vision language models (VLMs) and inspects how these models encode protected identities while generating images. We show that when VLMs are requested to transform human face images to reflect race identities (e.g., Hispanic, Black, White, or Asian) and sexual orientation (e.g., gay or straight), VLMs often comply with such requests. We demonstrate that compliance (and refusal) is non-deterministic, group-sensitive, and generated images are statistically distinguishable, often accentuating deep-seated stereotypes. We further show that representation biases introduced during image generation can re-emerge in textual reasoning and characterization tasks. Finally, we show that this worrisome finding is not limited to racial identity or sexual orientation, image generation can perpetuate bias around criminality, raising concerning possibilities of AI-assisted profiling.
| Comments: | Accepted at EMNLP 2026 (Main Conference Track) |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI); Computation and Language (cs.CL); Computers and Society (cs.CY) |
| Cite as: | arXiv:2610.35910 [cs.CV] |
| (or arXiv:2610.35910v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.35910 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Ashiqur Rahman KhudaBukhsh [view email]
[v1]
Tue, 6 Oct 2026 17:00:47 UTC (44,868 KB)
来源：arXiv cs.CY · arxiv.org