Evaluating Demographic Misrepresentation in Image-to-Image Portrait Editing

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Seo, Huichan, Hong, Minki, Choi, Sieun, Kim, Jihie, Oh, Jean
Format: Preprint
Veröffentlicht: 2026
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866914336258129920
author Seo, Huichan
Hong, Minki
Choi, Sieun
Kim, Jihie
Oh, Jean
author_facet Seo, Huichan
Hong, Minki
Choi, Sieun
Kim, Jihie
Oh, Jean
contents 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
format Preprint
id arxiv_https___arxiv_org_abs_2602_16149
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle Evaluating Demographic Misrepresentation in Image-to-Image Portrait Editing
Seo, Huichan
Hong, Minki
Choi, Sieun
Kim, Jihie
Oh, Jean
Computer Vision and Pattern Recognition
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
title Evaluating Demographic Misrepresentation in Image-to-Image Portrait Editing
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2602.16149