Insights from a Disaggregated Analysis of Kinds of Biases in a Multicultural Dataset
Published in Proceedings of the 9th Widening NLP Workshop (WiNLP 2025), 2025
Aggregated bias metrics often conceal critical disparities across intersectional demographic categories. This study provides a disaggregated analysis of stereotypes across multicultural data, showing how different bias categories manifest unevenly and emphasizing the importance of fine-grained evaluation.