Personalized Safety Alignment for Text-to-Image Diffusion 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_ | 1866912880157261824 |
|---|---|
| author | Lei, Yu Bai, Jinbin Shi, Qingyu Feng, Aosong Gao, Hongcheng Zhang, Xiao Ying, Rex |
| author_facet | Lei, Yu Bai, Jinbin Shi, Qingyu Feng, Aosong Gao, Hongcheng Zhang, Xiao Ying, Rex |
| contents | Text-to-image diffusion models have revolutionized visual content generation, yet their deployment is hindered by a fundamental limitation: safety mechanisms enforce rigid, uniform standards that fail to reflect diverse user preferences shaped by age, culture, or personal beliefs. To address this, we propose Personalized Safety Alignment (PSA), a framework that transitions generative safety from static filtration to user-conditioned adaptation. We introduce Sage, a large-scale dataset capturing diverse safety boundaries across 1,000 simulated user profiles, covering complex risks often missed by traditional datasets. By integrating these profiles via a parameter-efficient cross-attention adapter, PSA dynamically modulates generation to align with individual sensitivities. Extensive experiments demonstrate that PSA achieves a calibrated safety-quality trade-off: under permissive profiles, it relaxes over-cautious constraints to enhance visual fidelity, while under restrictive profiles, it enforces state-of-the-art suppression, significantly outperforming static baselines. Furthermore, PSA exhibits superior instruction adherence compared to prompt-engineering methods, establishing personalization as a vital direction for creating adaptive, user-centered, and responsible generative AI. Our code, data, and models are publicly available at https://github.com/M-E-AGI-Lab/PSAlign. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2508_01151 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Personalized Safety Alignment for Text-to-Image Diffusion Models Lei, Yu Bai, Jinbin Shi, Qingyu Feng, Aosong Gao, Hongcheng Zhang, Xiao Ying, Rex Computer Vision and Pattern Recognition Artificial Intelligence Text-to-image diffusion models have revolutionized visual content generation, yet their deployment is hindered by a fundamental limitation: safety mechanisms enforce rigid, uniform standards that fail to reflect diverse user preferences shaped by age, culture, or personal beliefs. To address this, we propose Personalized Safety Alignment (PSA), a framework that transitions generative safety from static filtration to user-conditioned adaptation. We introduce Sage, a large-scale dataset capturing diverse safety boundaries across 1,000 simulated user profiles, covering complex risks often missed by traditional datasets. By integrating these profiles via a parameter-efficient cross-attention adapter, PSA dynamically modulates generation to align with individual sensitivities. Extensive experiments demonstrate that PSA achieves a calibrated safety-quality trade-off: under permissive profiles, it relaxes over-cautious constraints to enhance visual fidelity, while under restrictive profiles, it enforces state-of-the-art suppression, significantly outperforming static baselines. Furthermore, PSA exhibits superior instruction adherence compared to prompt-engineering methods, establishing personalization as a vital direction for creating adaptive, user-centered, and responsible generative AI. Our code, data, and models are publicly available at https://github.com/M-E-AGI-Lab/PSAlign. |
| title | Personalized Safety Alignment for Text-to-Image Diffusion Models |
| topic | Computer Vision and Pattern Recognition Artificial Intelligence |
| url | https://arxiv.org/abs/2508.01151 |