Auditing Gender Presentation Differences in Text-to-Image Models
Open Access
- 29 October 2024
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 1-10
Abstract
Text-to-image models, which can generate high-quality images based on textual input, have recently enabled various content-creation tools. Despite significantly affecting a wide range of downstream applications, the distributions of these generated images are still not fully understood, especially when it comes to the potential stereotypical attributes of different genders. In this work, we propose a paradigm (Gender Presentation Differences) that utilizes fine-grained self-presentation attributes to study how gender is presented differently in text-to-image models. By probing gender indicators in the input text (e.g., “a woman” or “a man”), we quantify the frequency differences of presentation-centric attributes (e.g., “a shirt” and “a dress”) through human annotation and introduce a novel metric: GEP.1 Furthermore, we propose an automatic method to estimate such differences. The automatic GEP metric based on our approach yields a higher correlation with human annotations than that based on existing CLIP scores, consistently across three state-of-the-art text-to-image models. Finally, we demonstrate the generalization ability of our metrics in the context of gender stereotypes. We will publicly release our code/data.Keywords
All Related Versions
Funding Information
- Google Research Collabs program
This publication has 15 references indexed in Scilit:
- DALL-EVAL: Probing the Reasoning Skills and Social Biases of Text-to-Image Generation Models2023
- InstructPix2Pix: Learning to Follow Image Editing Instructions2023
- RePaint: Inpainting using Denoising Diffusion Probabilistic Models2022
- High-Resolution Image Synthesis with Latent Diffusion Models2022
- A Step Toward More Inclusive People Annotations for Fairness2021
- Gender Classification and Bias Mitigation in Facial Images2020
- How Computers See GenderProceedings of the ACM on Human-Computer Interaction, 2019
- Microsoft COCO: Common Objects in Context2014
- Comparison of the predicted and observed secondary structure of T4 phage lysozymeBiochimica et Biophysica Acta (BBA) - Protein Structure, 1975
- A NEW MEASURE OF RANK CORRELATIONBiometrika, 1938