When Detectors Forget Forensics: Blocking Semantic Shortcuts for Generalizable AI-Generated Image Detection

TL;DR

This paper introduces a novel method called Geometric Semantic Decoupling (GSD) to improve the generalization of AI-generated image detectors by removing reliance on semantic priors, thereby enhancing robustness against unseen generation methods.

Contribution

The paper identifies semantic fallback as a key failure in VFM-based detectors and proposes GSD, a parameter-free module that explicitly removes semantic components to improve detection robustness.

Findings

GSD significantly improves cross-dataset detection performance.

The method enhances robustness to unseen manipulations.

It generalizes beyond faces to general scene images.

Abstract

AI-generated image detection has become increasingly important with the rapid advancement of generative AI. However, detectors built on Vision Foundation Models (VFMs, \emph{e.g.}, CLIP) often struggle to generalize to images created using unseen generation pipelines. We identify, for the first time, a key failure mechanism, termed \emph{semantic fallback}, where VFM-based detectors rely on dominant pre-trained semantic priors (such as identity) rather than forgery-specific traces under distribution shifts. To address this issue, we propose \textbf{Geometric Semantic Decoupling (GSD)}, a parameter-free module that explicitly removes semantic components from learned representations by leveraging a frozen VFM as a semantic guide with a trainable VFM as an artifact detector. GSD estimates semantic directions from batch-wise statistics and projects them out via a geometric constraint, forcing the artifact detector to rely on semantic-invariant forensic evidence. Extensive experiments demonstrate that our method consistently outperforms state-of-the-art approaches, achieving 94.4\% video-level AUC (+\textbf{1.2\%}) in cross-dataset evaluation, improving robustness to unseen manipulations (+\textbf{3.0\%} on DF40), and generalizing beyond faces to the detection of synthetic images of general scenes, including UniversalFakeDetect (+\textbf{0.9\%}) and GenImage (+\textbf{1.7\%}).