Lune

EMNLP2025顶会

Exploring Changes in Nation Perception with Nationality-Assigned Personas in LLMs

Mahammed Kamruzzaman, Gene Louis Kim

摘要

Persona assignment has become a common strategy for customizing LLM use to particular tasks and contexts. In this study, we explore how evaluation of different nations changes when LLMs are assigned specific nationality personas. We assign 193 different nationality personas (e.g., an American person) to five LLMs and examine how the LLM evaluations (or "perceptions") of countries change. We find that all LLM-persona combinations tend to favor Western European nations, though nationpersonas push LLM behaviors to focus more on and treat the nation-persona's own region more favorably. Eastern European, Latin American, and African nations are treated more negatively by different nationality personas. We additionally find that evaluations by nationpersona LLMs of other nations correlate with human survey responses but fail to match the values closely. Our study provides insight into how biases and stereotypes are realized within LLMs when adopting different national personas. Our findings underscore the critical need for developing mechanisms to ensure that LLM outputs promote fairness and avoid overgeneralization. 1

问 Lune

尽管问,回答会引用原文。

Lune 会读完全文和相关的 Paper,每个回答都附上所依据的原文段落。

可以从这些问题问起

Lune 的依据

全文 · 引用它的 2 篇顶会 Paper · 相关的 Paper

获取答案

免费开始,无需绑卡

引用它的顶会 Paper2

  1. SurveyPilot: an Agentic Framework for Automated Human Opinion Collection from Social MediaViet Thanh Pham, Lizhen Qu, Zhuang Li, Suraj Sharma 等ACL 2025 · 4 次引用
  2. "As Eastern Powers, I Will Veto.": An Investigation of Nation-Level Bias of Large Language Models in International RelationsJonghyeon Choi, Yeonjun Choi, Hyun-chul Kim, Beakcheol JangAAAI 2026

它引用的 Paper15

  1. Generative Agents: Interactive Simulacra of Human BehaviorJoon Sung Park, Joseph C. O'Brien, Carrie Jun Cai, Meredith Ringel Morris 等UIST 2023 · 1,882 次引用
  2. Whose Opinions Do Language Models Reflect?Shibani Santurkar, Esin Durmus, Faisal Ladhak, Cinoo Lee 等ICML 2023 · 764 次引用
  3. Using Large Language Models to Simulate Multiple Humans and Replicate Human Subject StudiesGati V. Aher, Rosa I. Arriaga, Adam Tauman KalaiICML 2023 · 651 次引用
  4. SOTOPIA: Interactive Evaluation for Social Intelligence in Language AgentsXuhui Zhou, Hao Zhu, Leena Mathur, Ruohong Zhang 等ICLR 2024 · 288 次引用
  5. Using an LLM to Help With Code UnderstandingDaye Nam, Andrew Macvean, Vincent J. Hellendoorn, Bogdan Vasilescu 等ICSE 2024 · 264 次引用

相关 Paper

  1. Deus Ex Machina and Personas from Large Language Models: Investigating the Composition of AI-Generated Persona DescriptionsJoni Salminen, Chang Liu, Wenjing Pian, Jianxing Chi 等CHI 2024 · 55 次引用
  2. From Single to Societal: Analyzing Persona-Induced Bias in Multi-Agent InteractionsJiayi Li, Xiao Liu, Yansong FengAAAI 2026 · 3 次引用
  3. One Persona, Many Cues, Different Results: How Sociodemographic Cues Impact LLM PersonalizationFranziska Weeber, Vera Neplenbroek, Jan Batzner, Sebastian PadóACL 2026 · 4 次引用
  4. Missing the Margins: A Systematic Literature Review on the Demographic Representativeness of LLMsIndira Sen, Marlene Lutz, Elisa Rogers, David García 等ACL Findings 2025 · 2 次引用
  5. Two Tales of Persona in LLMs: A Survey of Role-Playing and PersonalizationYu-Min Tseng, Yu-Chao Huang, Teng-Yun Hsiao, Wei-Lin Chen 等EMNLP Findings 2024 · 87 次引用
黄昏的海面,两侧是细线勾勒的悬崖