检测文生图模型中物体的刻板印象，发现默认输出偏向白人中年人。
When Cars Have Stereotypes: Auditing Demographic Bias in Objects from Text-to-Image Models
- 提出SODA框架，自动发现属性并用三个指标量化偏见。
- 26.6%组合生成的20张图属性完全一致，如女性配玫瑰金笔记本。
- 去偏后组间差异减小，但组内多样性消失，新刻板印象取代旧的。
尽管先前研究多关注人类图像中的偏见，物体生成中的种族偏见仍较少被探讨。本文提出SODA（刻板物体诊断审计）框架，通过自动化属性发现与三项标准指标——基础与群体差异（BDS）、跨群体差异（CDS）、视觉属性集中度（VAC），系统测量此类偏见。在五种先进模型、八类物体（如汽车）上生成的8,000张图像中，发现‘中性’提示下输出最接近中年白人，暗示模型默认代表这些群体。此外，包含人口统计线索的提示导致高度刻板化输出：26.6%的对象-模型-群体组合中，20张生成图属性完全相同（如女性全为玫瑰金笔记本）。最后，提示层面的去偏策略虽降低组间差异，却导致组内多样性崩溃，以新刻板印象替代旧有偏见。SODA提供可量化的隐性关联分析路径，推动更负责任的AI发展。
原文摘要 · Abstract (English)
While prior research on text-to-image generation has predominantly focused on biases in human depictions, demographic bias in generated objects remains relatively underexplored. We introduce SODA (Stereotyped Object Diagnostic Audit), a novel framework for systematically measuring these biases through automated attribute discovery and three standardized metrics: Base vs. Demographic Divergence (BDS), Cross-Demographic Disparity (CDS), and Visual Attribute Concentration (VAC). Applying SODA to 8,000 images across five state-of-the-art models and eight object categories (e.g., cars), we find that "neutral" prompts produce outputs most visually similar to middle-aged and White people, suggesting these groups are implicitly over-represented in model defaults. Furthermore, demographic cues trigger highly skewed stereotypical outputs: 26.6% of object-model-demographic combinations produce results where all 20 generated images share the exact same attribute value (e.g., rose gold laptops for women). Finally, prompt-level debiasing reduces inter-group disparity but paradoxically collapses within-group diversity, replacing one stereotype with another. SODA offers a practical pipeline for making these implicit associations measurable, serving as a step toward more responsible AI development.
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。