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

Online social media platforms increasingly rely on Natural LanguageProcessing (NLP) techniques to detect abusive content at scale in order tomitigate the harms it causes to their users. However, these techniques sufferfrom various sampling and association biases present in training data, oftenresulting in sub-par performance on content relevant to marginalized groups,potentially furthering disproportionate harms towards them. Studies on suchbiases so far have focused on only a handful of axes of disparities andsubgroups that have annotations/lexicons available. Consequently, biasesconcerning non-Western contexts are largely ignored in the literature. In thispaper, we introduce a weakly supervised method to robustly detect lexicalbiases in broader geocultural contexts. Through a case study on a publiclyavailable toxicity detection model, we demonstrate that our method identifiessalient groups of cross-geographic errors, and, in a follow up, demonstratethat these groupings reflect human judgments of offensive and inoffensivelanguage in those geographic contexts. We also conduct analysis of a modeltrained on a dataset with ground truth labels to better understand thesebiases, and present preliminary mitigation experiments.

Detecting Cross-Geographic Biases in Toxicity Modeling on Social Media