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

Existing textual adversarial attacks usually utilize the gradient orprediction confidence to generate adversarial examples, making it hard to bedeployed in real-world applications. To this end, we consider a rarelyinvestigated but more rigorous setting, namely hard-label attack, in which theattacker can only access the prediction label. In particular, we find we canlearn the importance of different words via the change on prediction labelcaused by word substitutions on the adversarial examples. Based on thisobservation, we propose a novel adversarial attack, termed Text Hard-labelattacker (TextHacker). TextHacker randomly perturbs lots of words to craft anadversarial example. Then, TextHacker adopts a hybrid local search algorithmwith the estimation of word importance from the attack history to minimize theadversarial perturbation. Extensive evaluations for text classification andtextual entailment show that TextHacker significantly outperforms existinghard-label attacks regarding the attack performance as well as adversaryquality.

TextHacker: Learning based Hybrid Local Search Algorithm for Text Hard-label Adversarial Attack