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

We introduce HATELEXICON, a lexicon of slurs and targets of hate speech forthe countries of Brazil, Germany, India and Kenya, to aid training andinterpretability of models. We demonstrate how our lexicon can be used tointerpret model predictions, showing that models developed to classify extremespeech rely heavily on target words when making predictions. Further, wepropose a method to aid shot selection for training in low-resource settingsvia HATELEXICON. In few-shot learning, the selection of shots is of paramountimportance to model performance. In our work, we simulate a few-shot settingfor German and Hindi, using HASOC data for training and the MultilingualHateCheck (MHC) as a benchmark. We show that selecting shots based on ourlexicon leads to models performing better on MHC than models trained on shotssampled randomly. Thus, when given only a few training examples, using ourlexicon to select shots containing more sociocultural information leads tobetter few-shot performance.

Sociocultural knowledge is needed for selection of shots in hate speech detection tasks