发现图像编辑模型会无意识改变人脸肤色，尤其对非白人面孔影响更大。
Toward Trustworthy Portrait Editing: Evaluation of Demographic Misrepresentation in I2I Models
- 通过控制实验检测编辑模型在人物画像中对肤色等身份特征的隐性偏移
- 72%~75%的非白人面孔被错误变浅，而白人面孔仅44%受影响
- 添加明确外貌描述可有效减少非白人面孔的肤色变化，无需修改模型
指令引导的图像到图像（I2I）编辑器在消费与专业视觉流程中日益普及，其可信度不仅取决于指令遵循，还依赖于身份相关属性的公平保留。我们定义了两种失效模式：软抹除（请求编辑弱化或被沉默抑制）和刻板替换（引入未请求但符合刻板印象的族裔特征）。基于5,040张受控编辑肖像的基准测试，我们使用视觉-语言模型评分与人工评估，检验三种近期开源权重编辑器的表现。结果表明，身份保留失败普遍存在且具有种族不均衡性：62%~71%的输出出现皮肤变浅，其中印度及黑人源图像受影响达72%~75%，白人源图像仅为44%，显示在身份约束不足时输出倾向更浅或更白化的外观。在缓解案例研究中，添加提示级外观约束可使非白人源图像的种族变化得分降低最多1.48点，而白人源图像基本不变，无需修改模型权重。这些发现表明，身份保留并非I2I肖像编辑系统的普遍属性，而是分布不均的信任缺陷，具有直接社会后果。在部署规模下，此类无声畸变可能塑造人工智能中介的自我呈现并强化表征不平等。我们提出一种公平感知的审计协议，用于生成式编辑系统的评估与治理。
原文摘要 · Abstract (English)
Instruction-guided image-to-image (I2I) editors are increasingly used in consumer and professional visual workflows, where trustworthiness depends not only on prompt compliance but also on equitable preservation of identity-relevant attributes. We formalize two failure modes: Soft Erasure, where requested edits are weakly realized or silently suppressed, and Stereotype Replacement, where edits introduce unrequested, stereotype-consistent demographic attributes. Using a controlled benchmark of 5,040 edited portraits, we evaluate these failures across three recent open-weight editors with vision-language model scoring and human evaluation. Our results show that identity-preservation failures are pervasive and demographically uneven. In particular, 62--71% of outputs exhibit skin lightening, with Indian and Black source portraits affected at 72--75%, compared with 44% for White source portraits, indicating output-level drift toward lighter or more White-presenting appearances when identity constraints are underspecified. In a mitigation case study, prompt-level appearance constraints reduce race-change scores for non-White source portraits by up to 1.48 points, while leaving White source portraits largely unchanged, without modifying model weights. These findings show that identity preservation is not a uniform property of I2I portrait editing systems, but an unevenly distributed trustworthiness failure with direct social consequences. At deployment scale, such silent distortions can shape AI-mediated self-representation and reinforce representational disparities. We introduce a controlled audit protocol for fairness-aware evaluation and governance of generative editing systems. Project page: https://seochan99.github.io/i2i-demographic-bias
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。