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

An adversarial example is an input transformed by small perturbations thatmachine learning models consistently misclassify. While there are a number ofmethods proposed to generate adversarial examples for text data, it is nottrivial to assess the quality of these adversarial examples, as minorperturbations (such as changing a word in a sentence) can lead to a significantshift in their meaning, readability and classification label. In this paper, wepropose an evaluation framework consisting of a set of automatic evaluationmetrics and human evaluation guidelines, to rigorously assess the quality ofadversarial examples based on the aforementioned properties. We experiment withsix benchmark attacking methods and found that some methods generateadversarial examples with poor readability and content preservation. We alsolearned that multiple factors could influence the attacking performance, suchas the length of the text inputs and architecture of the classifiers.

Elephant in the Room: An Evaluation Framework for Assessing Adversarial Examples in NLP