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

Human biases are ubiquitous but not uniform: disparities exist acrosslinguistic, cultural, and societal borders. As large amounts of recentliterature suggest, language models (LMs) trained on human data can reflect andoften amplify the effects of these social biases. However, the vast majority ofexisting studies on bias are heavily skewed towards Western and Europeanlanguages. In this work, we scale the Word Embedding Association Test (WEAT) to24 languages, enabling broader studies and yielding interesting findings aboutLM bias. We additionally enhance this data with culturally relevant informationfor each language, capturing local contexts on a global scale. Further, toencompass more widely prevalent societal biases, we examine new bias dimensionsacross toxicity, ableism, and more. Moreover, we delve deeper into the Indianlinguistic landscape, conducting a comprehensive regional bias analysis acrosssix prevalent Indian languages. Finally, we highlight the significance of thesesocial biases and the new dimensions through an extensive comparison ofembedding methods, reinforcing the need to address them in pursuit of moreequitable language models. All code, data and results are available here:https://github.com/iamshnoo/weathub.

Global Voices, Local Biases: Socio-Cultural Prejudices across Languages