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 Tools annual

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 Tools monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Both generic and domain-specific BERT models are widely used for naturallanguage processing (NLP) tasks. In this paper we investigate the vulnerabilityof BERT models to variation in input data for Named Entity Recognition (NER)through adversarial attack. Experimental results show that BERT models arevulnerable to variation in the entity context with 20.2 to 45.0% of entitiespredicted completely wrong and another 29.3 to 53.3% of entities predictedwrong partially. BERT models seem most vulnerable to changes in the localcontext of entities and often a single change is sufficient to fool the model.The domain-specific BERT model trained from scratch (SciBERT) is morevulnerable than the original BERT model or the domain-specific model thatretains the BERT vocabulary (BioBERT). We also find that BERT models areparticularly vulnerable to emergent entities. Our results chart thevulnerabilities of BERT models for NER and emphasize the importance of furtherresearch into uncovering and reducing these weaknesses.

Breaking BERT: Understanding its Vulnerabilities for Named Entity Recognition through Adversarial Attack