aiqbench analysis
This study introduces the Stereotypes-to-Decisions (S2D) framework to systematically quantify how regional biases in large language models influence concrete social decisions regarding education, occupation, and interaction. By analyzing six LLMs across China's administrative regions, researchers found that these biases are systematic, correlate with economic indicators, and persist regardless of prompt language or specific model architecture. The findings underscore the urgent need for regionally aware evaluation protocols to mitigate harmful stereotypes before they impact real-world AI applications.
Evaluating Regional Bias in LLMs From Abstract Stereotype to Concrete Social Decision-Making
arXiv:2607.27022v1 Announce Type: new Abstract: Regional bias in large language models (LLMs) may shape both perceptions of regional groups and decisions about individuals from different regions. Yet existing studies often examine these manifestations separately, leaving their structure and consequences unclear. We introduce Stereotypes-to-Decisions (S2D), a systematic framework evaluating regional bias from abstract stereotypes to concrete social decisions. Covering all 34 provincial-level administrative regions of China, S2D evaluates six LLMs using stereotype ratings of Warmth (perceived friendliness and trustworthiness) and Competence (perceived capability and intelligence), along with paired-choice tasks across Education, Occupation, and Social Interaction. Results reveal substantial regional differences in regional scores, with considerable agreement across models, especially for Competence and Occupation decisions. Furthermore, these patterns are associated with regional economic and digital development indicators and display mixed human-like stereotypes, with some regions rated highly on one dimension but poorly on the other. They also remain largely stable across Chinese and English prompts. Overall, our findings show that regional bias in LLMs is prevalent, systematic, and consequential, motivating more regionally aware evaluation and mitigation.
Abstract:Regional bias in large language models (LLMs) may shape both perceptions of regional groups and decisions about individuals from different regions. Yet existing studies often examine these manifestations separately, leaving their structure and consequences unclear. We introduce Stereotypes-to-Decisions (S2D), a systematic framework evaluating regional bias from abstract stereotypes to concrete social decisions. Covering all 34 provincial-level administrative regions of China, S2D evaluates six LLMs using stereotype ratings of Warmth (perceived friendliness and trustworthiness) and Competence (perceived capability and intelligence), along with paired-choice tasks across Education, Occupation, and Social Interaction. Results reveal substantial regional differences in regional scores, with considerable agreement across models, especially for Competence and Occupation decisions. Furthermore, these patterns are associated with regional economic and digital development indicators and display mixed human-like stereotypes, with some regions rated highly on one dimension but poorly on the other. They also remain largely stable across Chinese and English prompts. Overall, our findings show that regional bias in LLMs is prevalent, systematic, and consequential, motivating more regionally aware evaluation and mitigation.
Computation and Language (cs.CL)
arXiv:2607.27022 [cs.CL]
(or arXiv:2607.27022v1 [cs.CL] for this version)
https://doi.org/10.48550/arXiv.2607.27022 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Full text under the publisher's open licence · 264 words
Related coverage
Which Values Do LLMs Confuse? A Schwartz-Based Recognition Study
arXiv · Jul 23, 2026
LLMs Contain Multitudes: How Deployment Context Reshapes Model-Level Preferences and Values
arXiv · Jun 15, 2026
The Expensive-Loop Problem
LessWrong · Oct 12, 2026
Podcast: The Future of AI: from Enterprise Adoption to Open Source Sovereignty
InfoQ · Sep 25, 2026