RA-Det: Towards Universal Detection of AI-Generated Images via Robustness Asymmetry

2026-03-02

AI Summary

This work proposes a behavior-driven, general-purpose framework for detecting AI-generated images, addressing the limited generalization of existing appearance-based methods when confronted with highly realistic synthetic content. The study reveals and theoretically analyzes a “robustness asymmetry” between natural and generated images: their semantic representations exhibit divergent stability under structured, minute perturbations. Notably, the approach requires no training data and is independent of generator-specific fingerprints. Detection is achieved through controlled perturbation probes, stability measurements in feature space, and theoretical lower-bound analysis. Extensive experiments across 14 generative models and over 10 state-of-the-art detectors demonstrate that the method outperforms current techniques by an average of 7.81%, exhibiting superior generalization and practical applicability.

Technology Category

• Computer Vision: Adversarial Attacks & Robustness
• Machine Learning: Adversarial Learning & Robustness
• Natural Language Processing: Safety and Robustness

Application Category

• Web Mining and Content Analysis: Robustness and generalizability of Web mining methods
• Social Networks and Social Media: Generative AI / large language models and their impact on social systems
• Economics, Online Markets and Human Computation: Economic ramifications for generative AI infrastructure and applications

Abstract

Recent image generators produce photo-realistic content that undermines the reliability of downstream recognition systems. As visual appearance cues become less pronounced, appearance-driven detectors that rely on forensic cues or high-level representations lose stability. This motivates a shift from appearance to behavior, focusing on how images respond to controlled perturbations rather than how they look. In this work, we identify a simple and universal behavioral signal. Natural images preserve stable semantic representations under small, structured perturbations, whereas generated images exhibit markedly larger feature drift. We refer to this phenomenon as robustness asymmetry and provide a theoretical analysis that establishes a lower bound connecting this asymmetry to memorization tendencies in generative models, explaining its prevalence across architectures. Building on this insight, we introduce Robustness Asymmetry Detection (RA-Det), a behavior-driven detection framework that converts robustness asymmetry into a reliable decision signal. Evaluated across 14 diverse generative models and against more than 10 strong detectors, RA-Det achieves superior performance, improving the average performance by 7.81 percent. The method is data- and model-agnostic, requires no generator fingerprints, and transfers across unseen generators. Together, these results indicate that robustness asymmetry is a stable, general cue for synthetic-image detection and that carefully designed probing can turn this cue into a practical, universal detector. The source code is publicly available at Github.

Problem

• AI-generated images
• synthetic-image detection
• robustness asymmetry
• universal detection
• image forensics

Innovation

• robustness asymmetry
• behavior-driven detection
• universal synthetic image detection
• feature drift
• model-agnostic detection