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

Pre-training by language modeling has become a popular and successfulapproach to NLP tasks, but we have yet to understand exactly what linguisticcapacities these pre-training processes confer upon models. In this paper weintroduce a suite of diagnostics drawn from human language experiments, whichallow us to ask targeted questions about the information used by languagemodels for generating predictions in context. As a case study, we apply thesediagnostics to the popular BERT model, finding that it can generallydistinguish good from bad completions involving shared category or rolereversal, albeit with less sensitivity than humans, and it robustly retrievesnoun hypernyms, but it struggles with challenging inferences and role-basedevent prediction -- and in particular, it shows clear insensitivity to thecontextual impacts of negation.

What BERT is not: Lessons from a new suite of psycholinguistic diagnostics for language models