Fragile Reconstruction: Adversarial Vulnerability of Reconstruction-Based Detectors for Diffusion-Generated Images
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915936686047232 |
|---|---|
| author | Jiang, Haoyang Yi, Mingyang Zhang, Shaolei Cai, Junxian Liu, Qingbin Chen, Xi Fan, Ju |
| author_facet | Jiang, Haoyang Yi, Mingyang Zhang, Shaolei Cai, Junxian Liu, Qingbin Chen, Xi Fan, Ju |
| contents | Recently, detecting AI-generated images produced by diffusion-based models has attracted increasing attention due to their potential threat to safety. Among existing approaches, reconstruction-based methods have emerged as a prominent paradigm for this task. However, we find that such methods exhibit severe security vulnerabilities to adversarial perturbations; that is, by adding imperceptible adversarial perturbations to input images, the detection accuracy of classifiers collapses to near zero. To verify this threat, we present a systematic evaluation of the adversarial robustness of three representative detectors across four diverse generative backbone models. First, we construct adversarial attacks in white-box scenarios, which degrade the performance of all well-trained detectors. Moreover, we find that these attacks demonstrate transferability; specifically, attacks crafted against one detector can be transferred to others, indicating that adversarial attacks on detectors can also be constructed in a black-box setting. Finally, we assess common countermeasures and find that standard defense methods against adversarial attacks provide limited mitigation. We attribute these failures to the low signal-to-noise ratio (SNR) of attacked samples as perceived by the detectors. Overall, our results reveal fundamental security limitations of reconstruction-based detectors and highlight the need to rethink existing detection strategies. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2604_12781 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | Fragile Reconstruction: Adversarial Vulnerability of Reconstruction-Based Detectors for Diffusion-Generated Images Jiang, Haoyang Yi, Mingyang Zhang, Shaolei Cai, Junxian Liu, Qingbin Chen, Xi Fan, Ju Computer Vision and Pattern Recognition Recently, detecting AI-generated images produced by diffusion-based models has attracted increasing attention due to their potential threat to safety. Among existing approaches, reconstruction-based methods have emerged as a prominent paradigm for this task. However, we find that such methods exhibit severe security vulnerabilities to adversarial perturbations; that is, by adding imperceptible adversarial perturbations to input images, the detection accuracy of classifiers collapses to near zero. To verify this threat, we present a systematic evaluation of the adversarial robustness of three representative detectors across four diverse generative backbone models. First, we construct adversarial attacks in white-box scenarios, which degrade the performance of all well-trained detectors. Moreover, we find that these attacks demonstrate transferability; specifically, attacks crafted against one detector can be transferred to others, indicating that adversarial attacks on detectors can also be constructed in a black-box setting. Finally, we assess common countermeasures and find that standard defense methods against adversarial attacks provide limited mitigation. We attribute these failures to the low signal-to-noise ratio (SNR) of attacked samples as perceived by the detectors. Overall, our results reveal fundamental security limitations of reconstruction-based detectors and highlight the need to rethink existing detection strategies. |
| title | Fragile Reconstruction: Adversarial Vulnerability of Reconstruction-Based Detectors for Diffusion-Generated Images |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2604.12781 |