Gender Artifacts from Art History to Text-to-Image Generation
For AI researchers and practitioners working on text-to-image generation, this work highlights how style keywords encode and amplify gender biases, providing a new dataset and metrics to study this issue.
The paper introduces StyleGender, a dataset of 74k images across 19 artistic styles, and proposes Set Gender Artifact (SGA) metrics to show that gender representation shapes visual features in both historical art and text-to-image generation, with generative models amplifying gender artifacts beyond historical sources.
Artistic styles are rooted in specific socio-historical contexts that encode social hierarchies, including distinct constructions of gender. Yet in AI research, style has long been treated as a surface-level visual property: a filter of color, brushstroke, and texture applied to otherwise content-neutral scenes. We introduce the first dataset to investigate the interplay between gender representation and style in both historical and generated images. StyleGender comprises 74k images spanning 19 artistic styles, comprising art historical images with style and gender annotations, T2I-generated images under controlled style and gender prompts, and a semantically aligned set enabling direct art history-to-generation comparison. By proposing two Set Gender Artifact (SGA) metrics (PixelSGA and MaskSGA), capturing gender signals at the pixel level and in compositional structure, we show that (1) gender representation shapes visual features across artistic styles, (2) style keywords carry these patterns into T2I generation, and (3) generative models tend to amplify gender artifacts beyond what is observed in historical sources.