# Detecting Adversarial Data using Perturbation Forgery Source: [Detecting Adversarial Data using Perturbation Forgery]() Collection index: [llms.txt]() Data updated: 2026-01-02T09:01:30.816Z Authors: Qian Wang, Chen Li, Yuchen Luo, Hefei Ling, Shijuan Huang, Ruoxi Jia, Ning Yu Published: 2024-05-25T13:34:16.000Z ## Abstract As a defense strategy against adversarial attacks, adversarial detection aims to identify and filter out adversarial data from the data flow based on discrepancies in distribution and noise patterns between natural and adversarial data. Although previous detection methods achieve high performance in detecting gradient-based adversarial attacks, new attacks based on generative models with imbalanced and anisotropic noise patterns evade detection. Even worse, the significant inference time overhead and limited performance against unseen attacks make existing techniques impractical for real-world use. In this paper, we explore the proximity relationship among adversarial noise distributions and demonstrate the existence of an open covering for these distributions. By training on the open covering of adversarial noise distributions, a detector with strong generalization performance against various types of unseen attacks can be developed. Based on this insight, we heuristically propose Perturbation Forgery, which includes noise distribution perturbation, sparse mask generation, and pseudo-adversarial data production, to train an adversarial detector capable of detecting any unseen gradient-based, generative-based, and physical adversarial attacks. Comprehensive experiments conducted on multiple general and facial datasets, with a wide spectrum of attacks, validate the strong generalization of our method. ## Recorded benchmark results A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage. - [Generative-based Adversarial Attack Detection on ImageNet100](): Generative-based adversarial attack detection; ImageNet100; CDA: 0.9878. - [Adversarial Attack Detection on Face dataset TIPIM attack](): Adversarial Attack Detection; Face dataset TIPIM attack; AUROC: 0.9885. - [Adversarial Attack Detection on Face dataset Adv-Sticker attack](): Adversarial Attack Detection; Face dataset Adv-Sticker attack; AUROC: 0.9999. - [Adversarial Attack Detection on Face dataset Adv-Glasses attack](): Adversarial Attack Detection; Face dataset Adv-Glasses attack; AUROC: 99.99. - [Adversarial Attack Detection on Face dataset Adv-Mask attack](): Adversarial Attack Detection; Face dataset Adv-Mask attack; AUROC: 99.99. - [Adversarial Attack Detection on Face dataset Adv-Makeup attack](): Adversarial Attack Detection; Face dataset Adv-Makeup attack; AUROC: 0.9762. - [Adversarial Attack Detection on Face dataset AMT-GAN attack](): Adversarial Attack Detection; Face dataset AMT-GAN attack; AUROC: 0.9216. - [Adversarial Detection on ImageNet MIFGSM attack (test)](): Adversarial Detection; ImageNet MIFGSM attack (test); AUROC: 99.31. - [Adversarial Attack Detection on ImageNet100](): Adversarial Attack Detection; ImageNet100; Robustness (BIM): 99.11. ## Original sources and implementation [Original paper]() [PDF]() [Implementation]()