Evaluating Demographic Misrepresentation in Image-to-Image Portrait Editing
This addresses demographic misrepresentation in AI systems for users of image editing tools, highlighting a central failure mode that motivates more robust editing systems.
The paper tackles the problem of demographic bias in image-to-image portrait editing by showing that identical edit instructions lead to systematically different outcomes across subject demographics, with failures like soft erasure and stereotype replacement being pervasive and demographically uneven. It demonstrates that a prompt-level identity constraint can substantially reduce demographic change for minority groups, revealing asymmetric identity priors in current editors.
Demographic bias in text-to-image (T2I) generation is well studied, yet demographic-conditioned failures in instruction-guided image-to-image (I2I) editing remain underexplored. We examine whether identical edit instructions yield systematically different outcomes across subject demographics in open-weight I2I editors. We formalize two failure modes: Soft Erasure, where edits are silently weakened or ignored in the output image, and Stereotype Replacement, where edits introduce unrequested, stereotype-consistent attributes. We introduce a controlled benchmark that probes demographic-conditioned behavior by generating and editing portraits conditioned on race, gender, and age using a diagnostic prompt set, and evaluate multiple editors with vision-language model (VLM) scoring and human evaluation. Our analysis shows that identity preservation failures are pervasive, demographically uneven, and shaped by implicit social priors, including occupation-driven gender inference. Finally, we demonstrate that a prompt-level identity constraint, without model updates, can substantially reduce demographic change for minority groups while leaving majority-group portraits largely unchanged, revealing asymmetric identity priors in current editors. Together, our findings establish identity preservation as a central and demographically uneven failure mode in I2I editing and motivate demographic-robust editing systems. Project page: https://seochan99.github.io/i2i-demographic-bias