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

Adversarial attacks and defenses have gained increasing interest on computervision systems in recent years, but as of today, most investigations arelimited to images. However, many artificial intelligence models actually handledocumentary data, which is very different from real world images. Hence, inthis work, we try to apply the adversarial attack philosophy on documentary andnatural data and to protect models against such attacks. We focus our work onuntargeted gradient-based, transfer-based and score-based attacks and evaluatethe impact of adversarial training, JPEG input compression and grey-scale inputtransformation on the robustness of ResNet50 and EfficientNetB0 modelarchitectures. To the best of our knowledge, no such work has been conducted bythe community in order to study the impact of these attacks on the documentimage classification task.

Evaluating Adversarial Robustness on Document Image Classification