mlsanityData to 7 Oct 2026

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

arXiv 2603.06264 ↗First posted 6 Mar 2026cs.CL cs.CYcited by 0

On the map: LLM alignment / RLHF / preference optimisation → Human values and AI feedback alignment

Abstract (arXiv)

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.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Human values and AI feedback alignment, one of 10 directions in LLM alignment / RLHF / preference optimisation. Builds on 19 of its 73 references that are in the map. First posted on arXiv 6 Mar 2026.

Builds on

19 of 73 references.

Closest ancestors · 1

Same direction, newest first

  1. 2024Investigating Cultural Alignment of Large Language Models20 Feb

Other references · 18

  1. 2021Ethical and social risks of harm from Language Modelscited by 957
  2. 2023Steering Language Models With Activation Engineeringcited by 824
  3. 2021BBQ: A Hand-Built Bias Benchmark for Question Answeringcited by 683
  4. 2023Towards Measuring the Representation of Subjective Global Opinions in Language Modelscited by 424
  5. 2024Bias in Large Language Models: Origin, Evaluation, and Mitigationcited by 77
  6. 2023Improving Activation Steering in Language Models with Mean-Centringcited by 64
  7. 2025Randomness, Not Representation: The Unreliability of Evaluating Cultural Alignment in LLMscited by 54
  8. 2024A Survey on Large Language Models with Multilingualism: Recent Advances and New Frontierscited by 39
  9. 2024Few-Shot Fairness: Unveiling LLM's Potential for Fairness-Aware Classificationcited by 18
  10. 2025Representation Engineering for Large-Language Models: Survey and Research Challengescited by 18
  11. 2023Watch Your Language: Investigating Content Moderation with Large Language Modelscited by 14
  12. 2025How Deep Is Representational Bias in LLMs? The Cases of Caste and Religioncited by 14
6 more
  1. 2025From Surveys to Narratives: Rethinking Cultural Value Adaptation in LLMscited by 9
  2. 2024Exploring Bengali Religious Dialect Biases in Large Language Models with Evaluation Perspectivescited by 4
  3. 2025Through the Prism of Culture: Evaluating LLMs' Understanding of Indian Subcultures and Traditionscited by 3
  4. 2025When Language Shapes Thought: Cross-Lingual Transfer of Factual Knowledge in Question Answeringcited by 2
  5. 2025Rethinking Hate Speech Detection on Social Media: Can LLMs Replace Traditional Models?cited by 2
  6. 2025Social Bias in Multilingual Language Models: A Surveycited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2025Multilingual != Multicultural: Evaluating Gaps Between Multilingual Capabilities and Cultural Alignment in LLMs
  2. 2026Understanding Cultural Alignment in Multilingual LLMs via Natural Debate Statements ↗
  3. 2025On the Alignment of Large Language Models with Global Human Opinion
  4. 2025Break the Checkbox: Challenging Closed-Style Evaluations of Cultural Alignment in LLMs
  5. 2026Can Persona-Prompted LLMs Emulate Subgroup Values? An Empirical Analysis of Generalisability and Fairness in Cultural Alignment
  6. 2023CDEval: A Benchmark for Measuring the Cultural Dimensions of Large Language Models
  7. 2023Cultural Alignment in Large Language Models: An Explanatory Analysis Based on Hofstede's Cultural Dimensions
  8. 2025Do Political Opinions Transfer Between Western Languages? An Analysis of Unaligned and Aligned Multilingual LLMs