Date of Award
Winter 3-14-2025
Document Type
Thesis (Undergraduate)
Department
Computer Science
First Advisor
Soroush Vosoughi
Abstract
Despite significant advancements in research on (intersectional) social biases in Large Language Models (LLMs), intersectional biases affecting subgroups within the LGBTQ+ community remain critically understudied. Existing bias detection methodologies often prioritize quantification but lack depth in explaining the specific stereotypes/biases that shape evaluation metrics. To address these gaps, this study proposes a combined statistical and visual-qualitative approach to quantify and identify persistent intersectional queer biases in five recent, state-of-the-art LLMs through a downstream story generation task. Findings from analysis uncover substantial evidence of stereotypes that perpetuate harmful, reductive narratives against intersectionally marginalized groups within the LGBTQ+ community. To promote public science and collaborative evaluation, the study also entails the publication of a web application, which enables users to interact with visualizations and contribute to qualitatively interpreting the biases present in the data.
Recommended Citation
Nguyen, Huu Duong (Chip), "Examining Intersectional Queer Biases in Large Language Models: A Combined Statistical and Visual-Qualitative Approach for Quantification and Explanation" (2025). Computer Science Senior Theses. 81.
https://digitalcommons.dartmouth.edu/cs_senior_theses/81
