测试中性提示下图像生成的性别肤色偏见，发现主流模型存在严重默认白人倾向。
Neutral Prompts, Non-Neutral People: Quantifying Gender and Skin-Tone Bias in Gemini Flash 2.5 Image and GPT Image 1.5
- 用四种中性提示生成3200张照片，结合肤色量化方法分析偏差
- 两模型输出超96%为白人，但性别偏好相反：Gemini偏女性，GPT偏男性
- 揭示中性提示实为探测工具，为视觉算法审计提供新方法
本研究量化了两款广泛部署的商用图像生成器——Gemini Flash 2.5 Image（NanoBanana）和GPT Image 1.5——在性别与肤色方面的偏见，检验了‘中性提示产生无偏输出’这一假设。通过四个语义中性的提示生成了3,200张写实风格图像。分析采用结合混合颜色归一化、面部关键点掩码及基于Monk (MST)、PERLA和Fitzpatrick量表的感知均匀肤色量化方法的严格流程。结果显示，中性提示并未产生无偏结果，反而导致强烈极化：两模型均表现出显著的“默认白人”倾向（输出中超过96%为白人）。但在性别方面，二者差异显著：Gemini倾向于女性特征主体，而GPT则更偏好具有浅肤色的男性特征主体。该研究通过光照感知的颜色度量方法，对前沿模型进行了大规模对比审计，区分了美学渲染与真实色素分布，证明中性提示实为诊断探针，而非中立指令。研究为审计算法视觉文化提供了可靠框架，并挑战了语言学上‘未标记即包容’的假设。
原文摘要 · Abstract (English)
This study quantifies gender and skin-tone bias in two widely deployed commercial image generators - Gemini Flash 2.5 Image (NanoBanana) and GPT Image 1.5 - to test the assumption that neutral prompts yield demographically neutral outputs. We generated 3,200 photorealistic images using four semantically neutral prompts. The analysis employed a rigorous pipeline combining hybrid color normalization, facial landmark masking, and perceptually uniform skin tone quantification using the Monk (MST), PERLA, and Fitzpatrick scales. Neutral prompts produced highly polarized defaults. Both models exhibited a strong "default white" bias (>96% of outputs). However, they diverged sharply on gender: Gemini favored female-presenting subjects, while GPT favored male-presenting subjects with lighter skin tones. This research provides a large-scale, comparative audit of state-of-the-art models using an illumination-aware colorimetric methodology, distinguishing aesthetic rendering from underlying pigmentation in synthetic imagery. The study demonstrates that neutral prompts function as diagnostic probes rather than neutral instructions. It offers a robust framework for auditing algorithmic visual culture and challenges the sociolinguistic assumption that unmarked language results in inclusive representation.
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。