Where Detectors Fail: Probing Generative Space for Generalizable AI-Generated Image Detection

2026-05-24

AI Summary

Current AI-generated image detectors exhibit limited generalization when confronted with unseen generative models or distribution shifts. To address this, this work proposes PROBE, a novel framework that repurposes the generator as a controllable exploration tool, leveraging detector feedback to actively synthesize challenging yet realistic instances directly on the image manifold. By integrating internal representation manipulation, manifold-level editing, and adversarial example generation, PROBE jointly optimizes the detector and generator, moving beyond conventional training paradigms reliant on static, large-scale datasets. Experimental results demonstrate that PROBE substantially enhances the detector’s generalization capability and robustness against previously unseen generative models.

Abstract

Detecting AI-generated images (AIGI) remains challenging because detectors often fail to generalize to unseen generators. Although existing methods are trained on large datasets, their performance still degrades when generation settings change, indicating that data scale alone is insufficient and that limited coverage of generative variations during training is a key factor. Studies on generative model editing show that small changes in internal representations can produce diverse and meaningful image variations, many of which are not explored under standard sampling. Leveraging this insight, we propose PROBE (Probing Robustness via Boundary Exploration), a framework that improves detector generalization by actively exploring challenging regions of the generative process. Instead of treating the generator as a fixed data source, PROBE uses the detector as a critic to steer the generator through manifold-level modifications, producing realistic samples that are difficult to classify. These samples expose failure cases that are uncommon under standard data sampling strategies and are used to refine the detector. Experimental results across multiple benchmarks indicate that PROBE enhances generalization to unseen generators, resulting in more generalizable AIGI detection performance. Code and models are available at https://github.com/Amamiya-C/PROBE-AIGI-Detection

Problem

• AI-generated image detection
• generalization
• unseen generators
• detector failure
• generative space

Innovation

• AI-generated image detection
• generalization
• generative space exploration
• manifold-level modification
• adversarial probing