Research

Paper

AI LLM March 06, 2026

Mind the Gap: Pitfalls of LLM Alignment with Asian Public Opinion

Authors

Hari Shankar, Vedanta S P, Sriharini Margapuri, Debjani Mazumder, Ponnurangam Kumaraguru, Abhijnan Chakraborty

Abstract

Large Language Models (LLMs) are increasingly being deployed in multilingual, multicultural settings, yet their reliance on predominantly English-centric training data risks misalignment with the diverse cultural values of different societies. In this paper, we present a comprehensive, multilingual audit of the cultural alignment of contemporary LLMs including GPT-4o-Mini, Gemini-2.5-Flash, Llama 3.2, Mistral and Gemma 3 across India, East Asia and Southeast Asia. Our study specifically focuses on the sensitive domain of religion as the prism for broader alignment. To facilitate this, we conduct a multi-faceted analysis of every LLM's internal representations, using log-probs/logits, to compare the model's opinion distributions against ground-truth public attitudes. We find that while the popular models generally align with public opinion on broad social issues, they consistently fail to accurately represent religious viewpoints, especially those of minority groups, often amplifying negative stereotypes. Lightweight interventions, such as demographic priming and native language prompting, partially mitigate but do not eliminate these cultural gaps. We further show that downstream evaluations on bias benchmarks (such as CrowS-Pairs, IndiBias, ThaiCLI, KoBBQ) reveal persistent harms and under-representation in sensitive contexts. Our findings underscore the urgent need for systematic, regionally grounded audits to ensure equitable global deployment of LLMs.

Metadata

arXiv ID: 2603.06264
Provider: ARXIV
Primary Category: cs.CL
Published: 2026-03-06
Fetched: 2026-03-09 06:05

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.06264v1</id>\n    <title>Mind the Gap: Pitfalls of LLM Alignment with Asian Public Opinion</title>\n    <updated>2026-03-06T13:29:54Z</updated>\n    <link href='https://arxiv.org/abs/2603.06264v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.06264v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>Large Language Models (LLMs) are increasingly being deployed in multilingual, multicultural settings, yet their reliance on predominantly English-centric training data risks misalignment with the diverse cultural values of different societies. In this paper, we present a comprehensive, multilingual audit of the cultural alignment of contemporary LLMs including GPT-4o-Mini, Gemini-2.5-Flash, Llama 3.2, Mistral and Gemma 3 across India, East Asia and Southeast Asia. Our study specifically focuses on the sensitive domain of religion as the prism for broader alignment. To facilitate this, we conduct a multi-faceted analysis of every LLM's internal representations, using log-probs/logits, to compare the model's opinion distributions against ground-truth public attitudes. We find that while the popular models generally align with public opinion on broad social issues, they consistently fail to accurately represent religious viewpoints, especially those of minority groups, often amplifying negative stereotypes. Lightweight interventions, such as demographic priming and native language prompting, partially mitigate but do not eliminate these cultural gaps. We further show that downstream evaluations on bias benchmarks (such as CrowS-Pairs, IndiBias, ThaiCLI, KoBBQ) reveal persistent harms and under-representation in sensitive contexts. Our findings underscore the urgent need for systematic, regionally grounded audits to ensure equitable global deployment of LLMs.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CY'/>\n    <published>2026-03-06T13:29:54Z</published>\n    <arxiv:comment>11 pages, including references</arxiv:comment>\n    <arxiv:primary_category term='cs.CL'/>\n    <author>\n      <name>Hari Shankar</name>\n    </author>\n    <author>\n      <name>Vedanta S P</name>\n    </author>\n    <author>\n      <name>Sriharini Margapuri</name>\n    </author>\n    <author>\n      <name>Debjani Mazumder</name>\n    </author>\n    <author>\n      <name>Ponnurangam Kumaraguru</name>\n    </author>\n    <author>\n      <name>Abhijnan Chakraborty</name>\n    </author>\n  </entry>"
}