FairImagen: Post-Processing for Bias Mitigation in Text-to-Image Models
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2025
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866917039715647488 |
|---|---|
| author | Fu, Zihao Brown, Ryan Shao, Shun Rawal, Kai Delaney, Eoin Russell, Chris |
| author_facet | Fu, Zihao Brown, Ryan Shao, Shun Rawal, Kai Delaney, Eoin Russell, Chris |
| contents | Text-to-image diffusion models, such as Stable Diffusion, have demonstrated remarkable capabilities in generating high-quality and diverse images from natural language prompts. However, recent studies reveal that these models often replicate and amplify societal biases, particularly along demographic attributes like gender and race. In this paper, we introduce FairImagen (https://github.com/fuzihaofzh/FairImagen), a post-hoc debiasing framework that operates on prompt embeddings to mitigate such biases without retraining or modifying the underlying diffusion model. Our method integrates Fair Principal Component Analysis to project CLIP-based input embeddings into a subspace that minimizes group-specific information while preserving semantic content. We further enhance debiasing effectiveness through empirical noise injection and propose a unified cross-demographic projection method that enables simultaneous debiasing across multiple demographic attributes. Extensive experiments across gender, race, and intersectional settings demonstrate that FairImagen significantly improves fairness with a moderate trade-off in image quality and prompt fidelity. Our framework outperforms existing post-hoc methods and offers a simple, scalable, and model-agnostic solution for equitable text-to-image generation. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_21363 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | FairImagen: Post-Processing for Bias Mitigation in Text-to-Image Models Fu, Zihao Brown, Ryan Shao, Shun Rawal, Kai Delaney, Eoin Russell, Chris Machine Learning Computation and Language Computer Vision and Pattern Recognition Text-to-image diffusion models, such as Stable Diffusion, have demonstrated remarkable capabilities in generating high-quality and diverse images from natural language prompts. However, recent studies reveal that these models often replicate and amplify societal biases, particularly along demographic attributes like gender and race. In this paper, we introduce FairImagen (https://github.com/fuzihaofzh/FairImagen), a post-hoc debiasing framework that operates on prompt embeddings to mitigate such biases without retraining or modifying the underlying diffusion model. Our method integrates Fair Principal Component Analysis to project CLIP-based input embeddings into a subspace that minimizes group-specific information while preserving semantic content. We further enhance debiasing effectiveness through empirical noise injection and propose a unified cross-demographic projection method that enables simultaneous debiasing across multiple demographic attributes. Extensive experiments across gender, race, and intersectional settings demonstrate that FairImagen significantly improves fairness with a moderate trade-off in image quality and prompt fidelity. Our framework outperforms existing post-hoc methods and offers a simple, scalable, and model-agnostic solution for equitable text-to-image generation. |
| title | FairImagen: Post-Processing for Bias Mitigation in Text-to-Image Models |
| topic | Machine Learning Computation and Language Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2510.21363 |