LKValues: Aligning Large Language Models with Sri Lankan Societal Values

2026-07-22

AI Summary

This study addresses the Western cultural bias prevalent in existing large language models, which impedes their alignment with Sri Lankan societal values and limits their evaluation capabilities within Sinhala-language contexts. Through a trilingual empirical survey involving 205 participants, the research integrates the global values framework with local cognitive perspectives to distill 40 core societal values. Leveraging this insight, the authors construct LKvaluesIT—a bilingual instruction-tuning corpus comprising 150,000 samples—and LKvaluesBench, a 1,000-sample evaluation benchmark. Fine-tuning the Qwen model on this resource suite yields significantly improved value alignment, reduced rates of invalid outputs, and diminished cross-lingual discrepancies in both English and Sinhala. This work establishes the first reproducible, nation-specific value alignment paradigm tailored for low-resource multilingual settings.

Abstract

Value alignment of Large Language Models (LLMs) has been shown to be culturally biased toward Western norms. This results in the mishandling of local values in multilingual societies such as Sri Lanka that have their unique cultural dynamics. Existing benchmarks overlook Sri Lankan-contextualized values in its official language Sinhala, hindering culturally sensitive evaluation and fine-tuning. To bridge this gap, we propose LKValues, the first survey-grounded resource suite for Sri Lankan value alignment. From a trilingual survey of 205 respondents, blending adapted global frameworks and LLM-elicited local constructs, we derive 40 majority-endorsed societal values. Using these values, we construct LKvaluesIT, a Sinhala-English news-derived instruction corpus containing 150k scenario-based instances, and LKvaluesBench, a value-sensitive evaluation benchmark of 1,000 instances. We evaluate a set of proprietary and open-weight LLMs with LKvaluesBench. We fine-tune three open-weight base models (Qwen3.5-4B-Base, Qwen3.5-9B-Base, and Aya-Expanse-8B-Base). Our experiments show that newer and larger LLMs still exhibit low-resource and cultural value-alignment gaps. LKValues fine-tuning improves Qwen-family models in English and Sinhala, reducing invalid outputs and cross-lingual disparities, though gains remain model-family dependent. These highlight LKValues efficacy in embedding Sri Lankan values, offering a replicable pipeline for low-resource, country-specific pluralist value alignment. The dataset is publicly available at https://github.com/NextME14/LKValues.

Problem

• value alignment
• cultural bias
• Sri Lankan values
• low-resource languages
• multilingual societies

Innovation

• value alignment
• low-resource languages
• Sinhala NLP
• culturally grounded AI
• instruction tuning