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

This paper proposes two intuitive metrics, skew and stereotype, that quantifyand analyse the gender bias present in contextual language models when tacklingthe WinoBias pronoun resolution task. We find evidence that gender stereotypecorrelates approximately negatively with gender skew in out-of-the-box models,suggesting that there is a trade-off between these two forms of bias. Weinvestigate two methods to mitigate bias. The first approach is an onlinemethod which is effective at removing skew at the expense of stereotype. Thesecond, inspired by previous work on ELMo, involves the fine-tuning of BERTusing an augmented gender-balanced dataset. We show that this reduces both skewand stereotype relative to its unaugmented fine-tuned counterpart. However, wefind that existing gender bias benchmarks do not fully probe professional biasas pronoun resolution may be obfuscated by cross-correlations from othermanifestations of gender prejudice. Our code is available online, athttps://github.com/12kleingordon34/NLP_masters_project.

Stereotype and Skew: Quantifying Gender Bias in Pre-trained and Fine-tuned Language Models