Mitigating the Risk of Health Inequity Exacerbated by Large Language Models

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Ji, Yuelyu, Ma, Wenhe, Sivarajkumar, Sonish, Zhang, Hang, Sadhu, Eugene Mathew, Li, Zhuochun, Wu, Xizhi, Visweswaran, Shyam, Wang, Yanshan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866929540116250624
author Ji, Yuelyu
Ma, Wenhe
Sivarajkumar, Sonish
Zhang, Hang
Sadhu, Eugene Mathew
Li, Zhuochun
Wu, Xizhi
Visweswaran, Shyam
Wang, Yanshan
author_facet Ji, Yuelyu
Ma, Wenhe
Sivarajkumar, Sonish
Zhang, Hang
Sadhu, Eugene Mathew
Li, Zhuochun
Wu, Xizhi
Visweswaran, Shyam
Wang, Yanshan
contents Recent advancements in large language models have demonstrated their potential in numerous medical applications, particularly in automating clinical trial matching for translational research and enhancing medical question answering for clinical decision support. However, our study shows that incorporating non decisive sociodemographic factors such as race, sex, income level, LGBT+ status, homelessness, illiteracy, disability, and unemployment into the input of LLMs can lead to incorrect and harmful outputs for these populations. These discrepancies risk exacerbating existing health disparities if LLMs are widely adopted in healthcare. To address this issue, we introduce EquityGuard, a novel framework designed to detect and mitigate the risk of health inequities in LLM based medical applications. Our evaluation demonstrates its efficacy in promoting equitable outcomes across diverse populations.
format Preprint
id arxiv_https___arxiv_org_abs_2410_05180
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Mitigating the Risk of Health Inequity Exacerbated by Large Language Models
Ji, Yuelyu
Ma, Wenhe
Sivarajkumar, Sonish
Zhang, Hang
Sadhu, Eugene Mathew
Li, Zhuochun
Wu, Xizhi
Visweswaran, Shyam
Wang, Yanshan
Computation and Language
Recent advancements in large language models have demonstrated their potential in numerous medical applications, particularly in automating clinical trial matching for translational research and enhancing medical question answering for clinical decision support. However, our study shows that incorporating non decisive sociodemographic factors such as race, sex, income level, LGBT+ status, homelessness, illiteracy, disability, and unemployment into the input of LLMs can lead to incorrect and harmful outputs for these populations. These discrepancies risk exacerbating existing health disparities if LLMs are widely adopted in healthcare. To address this issue, we introduce EquityGuard, a novel framework designed to detect and mitigate the risk of health inequities in LLM based medical applications. Our evaluation demonstrates its efficacy in promoting equitable outcomes across diverse populations.
title Mitigating the Risk of Health Inequity Exacerbated by Large Language Models
topic Computation and Language
url https://arxiv.org/abs/2410.05180