English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

In the evasion attacks against deep neural networks (DNN), the attackergenerates adversarial instances that are visually indistinguishable from benignsamples and sends them to the target DNN to trigger misclassifications. In thispaper, we propose a novel multi-view adversarial image detector, namely Argos,based on a novel observation. That is, there exist two "souls" in anadversarial instance, i.e., the visually unchanged content, which correspondsto the true label, and the added invisible perturbation, which corresponds tothe misclassified label. Such inconsistencies could be further amplifiedthrough an autoregressive generative approach that generates images with seedpixels selected from the original image, a selected label, and pixeldistributions learned from the training data. The generated images (i.e., the"views") will deviate significantly from the original one if the label isadversarial, demonstrating inconsistencies that Argos expects to detect. Tothis end, Argos first amplifies the discrepancies between the visual content ofan image and its misclassified label induced by the attack using a set ofregeneration mechanisms and then identifies an image as adversarial if thereproduced views deviate to a preset degree. Our experimental results show thatArgos significantly outperforms two representative adversarial detectors inboth detection accuracy and robustness against six well-known adversarialattacks. Code is available at:https://github.com/sohaib730/Argos-Adversarial_Detection

Two Souls in an Adversarial Image: Towards Universal Adversarial Example Detection using Multi-view Inconsistency