LLMs Homogenize Values in Constructive Arguments on Value-Laden Topics

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Shahid, Farhana, Zhang, Stella, Vashistha, Aditya
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908782081081344
author Shahid, Farhana
Zhang, Stella
Vashistha, Aditya
author_facet Shahid, Farhana
Zhang, Stella
Vashistha, Aditya
contents Large language models (LLMs) are increasingly used to promote prosocial and constructive discourse online. Yet little is known about how these models negotiate and shape underlying values when reframing people's arguments on value-laden topics. We conducted experiments with 465 participants from India and the United States, who wrote comments on homophobic and Islamophobic threads, and reviewed human-written and LLM-rewritten constructive versions of these comments. Our analysis shows that LLM systematically diminishes Conservative values while elevating prosocial values such as Benevolence and Universalism. When these comments were read by others, participants opposing same-sex marriage or Islam found human-written comments more aligned with their values, whereas those supportive of these communities found LLM-rewritten versions more aligned with their values. These findings suggest that value homogenization in LLM-mediated prosocial discourse runs the risk of marginalizing conservative viewpoints on value-laden topics and may inadvertently shape the dynamics of online discourse.
format Preprint
id arxiv_https___arxiv_org_abs_2509_10637
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle LLMs Homogenize Values in Constructive Arguments on Value-Laden Topics
Shahid, Farhana
Zhang, Stella
Vashistha, Aditya
Human-Computer Interaction
Computers and Society
Large language models (LLMs) are increasingly used to promote prosocial and constructive discourse online. Yet little is known about how these models negotiate and shape underlying values when reframing people's arguments on value-laden topics. We conducted experiments with 465 participants from India and the United States, who wrote comments on homophobic and Islamophobic threads, and reviewed human-written and LLM-rewritten constructive versions of these comments. Our analysis shows that LLM systematically diminishes Conservative values while elevating prosocial values such as Benevolence and Universalism. When these comments were read by others, participants opposing same-sex marriage or Islam found human-written comments more aligned with their values, whereas those supportive of these communities found LLM-rewritten versions more aligned with their values. These findings suggest that value homogenization in LLM-mediated prosocial discourse runs the risk of marginalizing conservative viewpoints on value-laden topics and may inadvertently shape the dynamics of online discourse.
title LLMs Homogenize Values in Constructive Arguments on Value-Laden Topics
topic Human-Computer Interaction
Computers and Society
url https://arxiv.org/abs/2509.10637