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 Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

This work examines the vulnerability of multimodal (image + text) models toadversarial threats similar to those discussed in previous literature onunimodal (image- or text-only) models. We introduce realistic assumptions ofpartial model knowledge and access, and discuss how these assumptions differfrom the standard "black-box"/"white-box" dichotomy common in currentliterature on adversarial attacks. Working under various levels of these"gray-box" assumptions, we develop new attack methodologies unique tomultimodal classification and evaluate them on the Hateful Memes Challengeclassification task. We find that attacking multiple modalities yields strongerattacks than unimodal attacks alone (inducing errors in up to 73% of cases),and that the unimodal image attacks on multimodal classifiers we explored werestronger than character-based text augmentation attacks (inducing errors onaverage in 45% and 30% of cases, respectively).

Adversarial Evaluation of Multimodal Models under Realistic Gray Box Assumption