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 expose important blind spots of deep learning systems.While word- and sentence-level attack scenarios mostly deal with findingsemantic paraphrases of the input that fool NLP models, character-level attackstypically insert typos into the input stream. It is commonly thought that theseare easier to defend via spelling correction modules. In this work, we showthat both a standard spellchecker and the approach of Pruthi et al. (2019),which trains to defend against insertions, deletions and swaps, perform poorlyon the character-level benchmark recently proposed in Eger and Benz (2020)which includes more challenging attacks such as visual and phoneticperturbations and missing word segmentations. In contrast, we show that anuntrained iterative approach which combines context-independent character-levelinformation with context-dependent information from BERT's masked languagemodeling can perform on par with human crowd-workers from Amazon MechanicalTurk (AMT) supervised via 3-shot learning.

BERT-Defense: A Probabilistic Model Based on BERT to Combat Cognitively Inspired Orthographic Adversarial Attacks