Research

Paper

AI LLM March 23, 2026

Efficient Zero-Shot AI-Generated Image Detection

Authors

Ryosuke Sonoda, Ramya Srinivasan

Abstract

The rapid progress of text-to-image models has made AI-generated images increasingly realistic, posing significant challenges for accurate detection of generated content. While training-based detectors often suffer from limited generalization to unseen images, training-free approaches offer better robustness, yet struggle to capture subtle discrepancies between real and synthetic images. In this work, we propose a training-free AI-generated image detection method that measures representation sensitivity to structured frequency perturbations, enabling detection of minute manipulations. The proposed method is computationally lightweight, as perturbation generation requires only a single Fourier transform for an input image. As a result, it achieves one to two orders of magnitude faster inference than most training-free detectors.Extensive experiments on challenging benchmarks demonstrate the efficacy of our method over state-of-the-art (SoTA). In particular, on OpenFake benchmark, our method improves AUC by nearly $10\%$ compared to SoTA, while maintaining substantially lower computational cost.

Metadata

arXiv ID: 2603.21619
Provider: ARXIV
Primary Category: cs.CV
Published: 2026-03-23
Fetched: 2026-03-24 06:02

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.21619v1</id>\n    <title>Efficient Zero-Shot AI-Generated Image Detection</title>\n    <updated>2026-03-23T06:33:20Z</updated>\n    <link href='https://arxiv.org/abs/2603.21619v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.21619v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>The rapid progress of text-to-image models has made AI-generated images increasingly realistic, posing significant challenges for accurate detection of generated content. While training-based detectors often suffer from limited generalization to unseen images, training-free approaches offer better robustness, yet struggle to capture subtle discrepancies between real and synthetic images. In this work, we propose a training-free AI-generated image detection method that measures representation sensitivity to structured frequency perturbations, enabling detection of minute manipulations. The proposed method is computationally lightweight, as perturbation generation requires only a single Fourier transform for an input image. As a result, it achieves one to two orders of magnitude faster inference than most training-free detectors.Extensive experiments on challenging benchmarks demonstrate the efficacy of our method over state-of-the-art (SoTA). In particular, on OpenFake benchmark, our method improves AUC by nearly $10\\%$ compared to SoTA, while maintaining substantially lower computational cost.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CV'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.AI'/>\n    <published>2026-03-23T06:33:20Z</published>\n    <arxiv:primary_category term='cs.CV'/>\n    <author>\n      <name>Ryosuke Sonoda</name>\n    </author>\n    <author>\n      <name>Ramya Srinivasan</name>\n    </author>\n  </entry>"
}