SC-Pro: Training-Free Framework for Defending Unsafe Image Synthesis Attack
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915306810638336 |
|---|---|
| author | Park, Junha Hwang, Jaehui Ryu, Ian Park, Hyungkeun Kim, Jiyoon Lee, Jong-Seok |
| author_facet | Park, Junha Hwang, Jaehui Ryu, Ian Park, Hyungkeun Kim, Jiyoon Lee, Jong-Seok |
| contents | With advances in diffusion models, image generation has shown significant performance improvements. This raises concerns about the potential abuse of image generation, such as the creation of explicit or violent images, commonly referred to as Not Safe For Work (NSFW) content. To address this, the Stable Diffusion model includes several safety checkers to censor initial text prompts and final output images generated from the model. However, recent research has shown that these safety checkers have vulnerabilities against adversarial attacks, allowing them to generate NSFW images. In this paper, we find that these adversarial attacks are not robust to small changes in text prompts or input latents. Based on this, we propose SC-Pro (Spherical or Circular Probing), a training-free framework that easily defends against adversarial attacks generating NSFW images. Moreover, we develop an approach that utilizes one-step diffusion models for efficient NSFW detection (SC-Pro-o), further reducing computational resources. We demonstrate the superiority of our method in terms of performance and applicability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2501_05359 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | SC-Pro: Training-Free Framework for Defending Unsafe Image Synthesis Attack Park, Junha Hwang, Jaehui Ryu, Ian Park, Hyungkeun Kim, Jiyoon Lee, Jong-Seok Computer Vision and Pattern Recognition With advances in diffusion models, image generation has shown significant performance improvements. This raises concerns about the potential abuse of image generation, such as the creation of explicit or violent images, commonly referred to as Not Safe For Work (NSFW) content. To address this, the Stable Diffusion model includes several safety checkers to censor initial text prompts and final output images generated from the model. However, recent research has shown that these safety checkers have vulnerabilities against adversarial attacks, allowing them to generate NSFW images. In this paper, we find that these adversarial attacks are not robust to small changes in text prompts or input latents. Based on this, we propose SC-Pro (Spherical or Circular Probing), a training-free framework that easily defends against adversarial attacks generating NSFW images. Moreover, we develop an approach that utilizes one-step diffusion models for efficient NSFW detection (SC-Pro-o), further reducing computational resources. We demonstrate the superiority of our method in terms of performance and applicability. |
| title | SC-Pro: Training-Free Framework for Defending Unsafe Image Synthesis Attack |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2501.05359 |