Automated In-the-Wild Data Collection for Continual AI Generated Image Detection

TL;DR

This paper presents a data-centric continual learning framework that automatically collects in-the-wild data and incorporates generator-driven data to improve AI-generated image detection under evolving conditions.

Contribution

It introduces an automated, weakly supervised pipeline for dataset construction and demonstrates effective continual adaptation to new generative models.

Findings

Achieved +9.14% and +8% improvements in average accuracy on two detectors.

Showed that combining in-the-wild and generator-driven data enhances robustness.

Validated the approach through extensive experiments.

Abstract

The rapid advancement of generative Artificial Intelligence (AI) has introduced significant challenges for reliable AI-generated image detection. Existing detectors often suffer from performance degradation under distribution shifts and when encountering newly emerging generative models. In this work, we propose a data-centric continual adaptation framework for updating detectors in evolving environments. We show that both in-the-wild data and generator-driven data are essential for adapting detectors. We introduce an automated, weakly supervised pipeline for constructing in-the-wild datasets through fact-check article retrieval. Additionally, we demonstrate that incorporating even a small amount of generator-driven data during training enables effective adaptation to newly emerging models, while combining it with in-the-wild data within a continual learning framework enables robust adaptation and mitigates catastrophic forgetting. Extensive experiments on two state-of-the-art detectors show significant improvements of +9.14% and +8% in average accuracy, respectively.