@inproceedings{wan-chang-2025-male,
    title = "The Male {CEO} and the Female Assistant: Evaluation and Mitigation of Gender Biases in Text-To-Image Generation of Dual Subjects",
    author = "Wan, Yixin  and
      Chang, Kai-Wei",
    editor = "Che, Wanxiang  and
      Nabende, Joyce  and
      Shutova, Ekaterina  and
      Pilehvar, Mohammad Taher",
    booktitle = "Proceedings of the 63rd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
    month = jul,
    year = "2025",
    address = "Vienna, Austria",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/paragraph-normalization/2025.acl-long.449/",
    doi = "10.18653/v1/2025.acl-long.449",
    pages = "9174--9190",
    ISBN = "979-8-89176-251-0",
    abstract = "Recent large-scale T2I models like DALLE-3 have made progress in reducing gender stereotypes when generating single-person images. However, significant biases remain when generating images with more than one person. To systematically evaluate this, we propose the \textbf{Paired Stereotype Test (PST)} framework, which queries T2I models to depict two individuals assigned with male-stereotyped and female-stereotyped social identities, respectively (e.g. ``a CEO'' and ``an Assistant''). This contrastive setting often triggers T2I models to generate gender-stereotyped images. Using PST, we evaluate two aspects of gender biases {--} the well-known \textbf{bias in gendered occupation} and a novel aspect: \textbf{bias in organizational power}. Experiments show that \textbf{over 74{\%} images generated by DALLE-3 display gender-occupational biases}. Additionally, compared to single-person settings, DALLE-3 is more likely to perpetuate male-associated stereotypes under PST. We further propose \textbf{FairCritic}, a novel and interpretable framework that leverages an LLM-based critic model to i) detect bias in generated images, and ii) adaptively provide feedback to T2I models for improving fairness. FairCritic achieves near-perfect fairness on PST, overcoming the limitations of previous prompt-based intervention approaches."
}