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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

Models trained to estimate word probabilities in context have becomeubiquitous in natural language processing. How do these models use lexical cuesin context to inform their word probabilities? To answer this question, wepresent a case study analyzing the pre-trained BERT model with tests informedby semantic priming. Using English lexical stimuli that show priming in humans,we find that BERT too shows "priming," predicting a word with greaterprobability when the context includes a related word versus an unrelated one.This effect decreases as the amount of information provided by the contextincreases. Follow-up analysis shows BERT to be increasingly distracted byrelated prime words as context becomes more informative, assigning lowerprobabilities to related words. Our findings highlight the importance ofconsidering contextual constraint effects when studying word prediction inthese models, and highlight possible parallels with human processing.

Exploring BERT's Sensitivity to Lexical Cues using Tests from Semantic Priming