Exploiting the Final Component of Generator Architectures for AI-Generated Image Detection

TL;DR

This paper introduces a novel detection method that exploits shared final components in various image generators to distinguish AI-generated images from real ones, achieving high accuracy even on unseen generators.

Contribution

It proposes a new detection approach based on generator final components and provides a taxonomy for categorizing generators to improve generalization.

Findings

Achieves 98.83% average accuracy on unseen generators

Uses only 100 samples per category for training

Effective across diverse generator architectures

Abstract

With the rapid proliferation of powerful image generators, accurate detection of AI-generated images has become essential for maintaining a trustworthy online environment. However, existing deepfake detectors often generalize poorly to images produced by unseen generators. Notably, despite being trained under vastly different paradigms, such as diffusion or autoregressive modeling, many modern image generators share common final architectural components that serve as the last stage for converting intermediate representations into images. Motivated by this insight, we propose to "contaminate" real images using the generator's final component and train a detector to distinguish them from the original real images. We further introduce a taxonomy based on generators' final components and categorize 21 widely used generators accordingly, enabling a comprehensive investigation of our method's generalization capability. Using only 100 samples from each of three representative categories, our detector-fine-tuned on the DINOv3 backbone-achieves an average accuracy of 98.83% across 22 testing sets from unseen generators.