Proceedings

PURE: Purging Unrelated Representations for Content-Agnostic Forgery Detection

Xinyu Wu, Dong Li, Minglai Shao, Xintao Wu, Zhong Chen, Chen Zhao

Download BibTeX Cite

Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence. Main Track. Pages 1857-1865.

https://doi.org/10.24963/ijcai.2026/207

Keywords

  • Computer Vision: Machine learning for vision
  • Computer Vision: Multimodal learning
  • Computer Vision: Representation learning
  • Machine Learning: Causality
  • Machine Learning: Classification

Abstract

Existing AI-generated image (AIGI) detectors perform well in-domain but degrade severely under distribution shift. We observe that this failure is mainly caused by content shortcuts, where detectors spuriously couple forgery artifacts with semantic content, such as object categories or demographic attributes, learning content–label correlations instead of generalizable forgery patterns. To address this issue, we propose PURE (Purging Unrelated Representations for Content-Agnostic Forgery Detection), which achieves content-agnostic detection through two complementary components: a Causal Semantic Generative (CSG) mechanism that disentangles semantic representations from forgery-irrelevant nuisance factors, and a Gaussian Mixture Model (GMM)-based prototype alignment module that suppresses category-specific content bias. Extensive experiments on CIFAKE, GenImage, and AlFace show that PURE achieves superior generalization under spurious correlation reversal. The code is available at https://github.com/wuxinyu519/PURE.

Xinyu Wu, Dong Li, Minglai Shao, Xintao Wu, Zhong Chen, and Chen Zhao. 2026. PURE: Purging Unrelated Representations for Content-Agnostic Forgery Detection. In Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence, IJCAI-26. IJCAI, 1857–1865. https://doi.org/10.24963/ijcai.2026/207
Welcome Future Conferences AI Journal Division IJCAI Proceedings About IJCAI IJCAI Awards