Skip to main navigation Skip to search Skip to main content

PersonaGym: Evaluating Persona Agents and LLMs

  • Vinay Samuel
  • , Henry Peng Zou
  • , Yue Zhou
  • , Shreyas Chaudhari
  • , Ashwin Kalyan
  • , Tanmay Rajpurohit
  • , Ameet Deshpande
  • , Karthik Narasimhan
  • , Vishvak Murahari

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

Persona agents, which are LLM agents conditioned to act according to an assigned persona, enable contextually rich and user-aligned interactions across domains like education and healthcare. However, evaluating how faithfully these agents adhere to their personas remains a significant challenge, particularly in free-form settings that demand consistency across diverse, persona-relevant environments. We introduce PersonaGym, the first dynamic evaluation framework for persona agents, and PersonaScore, a human-aligned automatic metric grounded in decision theory that enables comprehensive large-scale evaluation. Our evaluation of 10 leading LLMs across 200 personas and 10,000 questions reveals significant advancement opportunities. For example, GPT-4.1 had the exact same PersonaScore as LLAMA-3-8B despite being a more recent and advanced closed-source model. Importantly, increased model size and complexity do not necessarily enhance persona agent capabilities, underscoring the need for algorithmic and architectural innovation toward faithful, performant persona agents.

Original languageEnglish (US)
Title of host publicationEMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025
EditorsChristos Christodoulopoulos, Tanmoy Chakraborty, Carolyn Rose, Violet Peng
PublisherAssociation for Computational Linguistics (ACL)
Pages6999-7022
Number of pages24
ISBN (Electronic)9798891763357
DOIs
StatePublished - 2025
Event30th Conference on Empirical Methods in Natural Language Processing, EMNLP 2025 - Suzhou, China
Duration: Nov 4 2025Nov 9 2025

Publication series

NameEMNLP 2025 - 2025 Conference on Empirical Methods in Natural Language Processing, Findings of EMNLP 2025

Conference

Conference30th Conference on Empirical Methods in Natural Language Processing, EMNLP 2025
Country/TerritoryChina
CitySuzhou
Period11/4/2511/9/25

All Science Journal Classification (ASJC) codes

  • Computational Theory and Mathematics
  • Computer Science Applications
  • Information Systems
  • Linguistics and Language

Fingerprint

Dive into the research topics of 'PersonaGym: Evaluating Persona Agents and LLMs'. Together they form a unique fingerprint.

Cite this