Year: 2025
Type: preprint
Abstract: The rapid advancement of generative models has made real and synthetic images increasingly indistinguishable. Although extensive efforts have been devoted to detecting AI-generated images, out-of-distribution generalization remains a persistent challenge. We trace this weakness to spurious shortcuts exploited during training and we also observe that small feature-space perturbations can mitigate shortcut dominance. To address this problem in a more controllable manner, we propose the Positive-Incentive Noise for CLIP (PiN-CLIP), which jointly trains a noise generator and a detection network under a variational positive-incentive principle. Specifically, we construct positive-incentive noise in the feature space via cross-attention fusion of visual and categorical semantic features. During optimization, the noise is injected into the feature space to fine-tune the visual encoder, suppressing shortcut-sensitive directions while amplifying stable forensic cues, thereby enabling the extraction of more robust and generalized artifact representations. Comparative experiments are conducted on an open-world dataset comprising synthetic images generated by 42 distinct generative models. Our method achieves new state-of-the-art performance, with notable improvements of 5.4 in average accuracy over existing approaches.
Source: arXiv (Cornell University)
Authors: Li, Ziqiang, Yan, Jiazhen, Wang, Fan, Zeng, Kai, Fu, Zhangjie
Subfield: Computer Vision and Pattern Recognition
Field: Computer Science
Domain: Physical Sciences
Open Access status: green
Awards: G331018939, G975555798
OpenAlex ID: W4416550355
Cites:
Cited by:
Related to: