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

We study an important and challenging task of attacking natural languageprocessing models in a hard label black box setting. We propose adecision-based attack strategy that crafts high quality adversarial examples ontext classification and entailment tasks. Our proposed attack strategyleverages population-based optimization algorithm to craft plausible andsemantically similar adversarial examples by observing only the top labelpredicted by the target model. At each iteration, the optimization procedureallow word replacements that maximizes the overall semantic similarity betweenthe original and the adversarial text. Further, our approach does not rely onusing substitute models or any kind of training data. We demonstrate theefficacy of our proposed approach through extensive experimentation andablation studies on five state-of-the-art target models across seven benchmarkdatasets. In comparison to attacks proposed in prior literature, we are able toachieve a higher success rate with lower word perturbation percentage that tooin a highly restricted setting.

Generating Natural Language Attacks in a Hard Label Black Box Setting