Evaluating Biases in Context-Dependent Sexual and Reproductive Health Questions
Sharon Levy, Tahilin Sanchez Karver, William Adler, Michelle R Kaufman, Mark Dredze
Abstract
Chat-based large language models have the opportunity to empower individuals lacking high-quality healthcare access to receive personalized information across a variety of topics. However, users may ask underspecified questions that require additional context for a model to correctly answer. We study how large language model biases are exhibited through these contextual questions in the healthcare domain. To accomplish this, we curate a dataset of sexual and reproductive healthcare questions (ContextSRH) that are dependent on age, sex, and location attributes. We compare models’ outputs with and without demographic context to determine answer alignment among our contextual questions. Our experiments reveal biases in each of these attributes, where young adult female users are favored.
BibTeX
@inproceedings{levy-etal-2024-evaluating,
title = "Evaluating Biases in Context-Dependent Sexual and Reproductive Health Questions",
author = "Levy, Sharon and
Karver, Tahilin Sanchez and
Adler, William and
Kaufman, Michelle R and
Dredze, Mark",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-emnlp.332/",
doi = "10.18653/v1/2024.findings-emnlp.332",
pages = "5801--5812"
}