Skip to main navigation Skip to search Skip to main content

Conformity, Confabulation, and Impersonation: Persona Inconstancy in Multi-Agent LLM Collaboration

  • Department of Electrical and Computer Engineering
  • University of Illinois at Urbana-Champaign

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

9 Scopus citations

Abstract

This study explores the sources of instability in maintaining cultural personas and opinions within multi-agent LLM systems. Drawing on simulations of inter-cultural collaboration and debate, we analyze agents’ pre- and post-discussion private responses alongside chat transcripts to assess the stability of cultural personas and the impact of opinion diversity on group outcomes. Our findings suggest that multi-agent discussions can encourage collective decisions that reflect diverse perspectives, yet this benefit is tempered by the agents’ susceptibility to conformity due to perceived peer pressure and challenges in maintaining consistent personas and opinions. Counterintuitively, instructions that encourage debate in support of one’s opinions increase the rate of inconstancy. Without addressing the factors we identify, the full potential of multi-agent frameworks for producing more culturally diverse AI outputs will remain untapped.

Original languageEnglish
Title of host publicationC3NLP 2024 - 2nd Workshop on Cross-Cultural Considerations in NLP, Proceedings of the Workshop
EditorsVinodkumar Prabhakaran, Sunipa Dev, Luciana Benotti, Daniel Hershcovich, Laura Cabello, Yong Cao, Ife Adebara, Li Zho
PublisherAssociation for Computational Linguistics (ACL)
Pages17-31
Number of pages15
ISBN (Electronic)9798891761469
StatePublished - 2024
Event2nd Workshop on Cross-Cultural Considerations in NLP, C3NLP 2024 - Bangkok, Thailand
Duration: Aug 16 2024 → …

Publication series

NameC3NLP 2024 - 2nd Workshop on Cross-Cultural Considerations in NLP, Proceedings of the Workshop

Conference

Conference2nd Workshop on Cross-Cultural Considerations in NLP, C3NLP 2024
Country/TerritoryThailand
CityBangkok
Period08/16/24 → …

Fingerprint

Dive into the research topics of 'Conformity, Confabulation, and Impersonation: Persona Inconstancy in Multi-Agent LLM Collaboration'. Together they form a unique fingerprint.

Cite this