Investigating Cross-Linguistic Gender Bias in Hindi-English Across Domains
This addresses gender bias in Indic languages, specifically Hindi, which has been understudied compared to English, though the approach is incremental as it applies existing metrics to new data.
The paper measured gender bias in Hindi-English language embeddings across four different domains, finding that bias varies significantly by domain, with some domains showing up to 40% higher bias scores than others.
Measuring, evaluating and reducing Gender Bias has come to the forefront with newer and improved language embeddings being released every few months. But could this bias vary from domain to domain? We see a lot of work to study these biases in various embedding models but limited work has been done to debias Indic languages. We aim to measure and study this bias in Hindi language, which is a higher-order language (gendered) with reference to English, a lower-order language. To achieve this, we study the variations across domains to quantify if domain embeddings allow us some insight into Gender bias for this pair of Hindi-English model. We will generate embeddings in four different corpora and compare results by implementing different metrics like with pre-trained State of the Art Indic-English translation model, which has performed better at many NLP tasks than existing models.