# Pulling Back the Curtain: Unsupervised Adversarial Detection via Contrastive Auxiliary Networks

Source: [Pulling Back the Curtain: Unsupervised Adversarial Detection via Contrastive Auxiliary Networks](<https://www.sota2.com/research/paper/68fa7a2c1e209131cea818c5>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2025-10-29T05:02:11.024Z

Authors: Eylon Mizrahi, Raz Lapid, Moshe Sipper

Published: 2025-02-13T09:40:26.000Z

## Abstract

Deep learning models are widely employed in safety-critical applications yet remain susceptible to adversarial attacks -- imperceptible perturbations that can significantly degrade model performance. Conventional defense mechanisms predominantly focus on either enhancing model robustness or detecting adversarial inputs independently. In this work, we propose an Unsupervised adversarial detection via Contrastive Auxiliary Networks (U-CAN) to uncover adversarial behavior within auxiliary feature representations, without the need for adversarial examples. U-CAN is embedded within selected intermediate layers of the target model. These auxiliary networks, comprising projection layers and ArcFace-based linear layers, refine feature representations to more effectively distinguish between benign and adversarial inputs. Comprehensive experiments across multiple datasets (CIFAR-10, Mammals, and a subset of ImageNet) and architectures (ResNet-50, VGG-16, and ViT) demonstrate that our method surpasses existing unsupervised adversarial detection techniques, achieving superior F1 scores against four distinct attack methods. The proposed framework provides a scalable and effective solution for enhancing the security and reliability of deep learning systems.

## 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.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2502.09110>)

[PDF](<https://arxiv.org/pdf/2502.09110.pdf>)