ニュース
LKValues: Aligning Large Language Models with Sri Lankan Societal Values
arXiv cs.AI · 公開日 · 読了3分
30秒で要点
- 何が起きたか
- Researchers created LKValues, a dataset and benchmark for aligning large language models with Sri Lankan cultural values, including 150k Sinhala-English training instances and evaluation benchmarks.
- なぜ重要か
- Engineers building or fine-tuning LLMs for South Asian markets or multilingual systems need culturally grounded alignment beyond Western-centric defaults.
- 注意点
- Fine-tuning improvements were model-family dependent and gaps remained even after training, suggesting cultural alignment requires ongoing iteration and context-specific approaches.
この要約を音声で聴く
記事より
-->
Computer Science > Computation and Language
arXiv:2607.20410v1 (cs)
[Submitted on 22 Jul 2026]
Title: LKValues: Aligning Large Language Models with Sri Lankan Societal Values
Authors: Nethmi Muthugala , Supryadi , Surangika Ranathunga , Nisansa de Silva , Ruijie Tao , Ovindu Gunatunga , Pengyun Zhu , Shaowei Zhang , Jingting Zheng , Deyi Xiong
View a PDF of the paper titled LKValues: Aligning Large Language Models with Sri Lankan Societal Values, by Nethmi Muthugala and 9 other authors
View PDF HTML (experimental)
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 evaluatio
原文からの抜粋です。全文は配信元でお読みください。
- llm
- language model
- fine-tun
The Agent Architect
1つのパターン、1つのトレードオフ、1つの本番障害事例。エージェントシステムを構築する人のための短い週刊ブリーフィング。
週1回のメール、ワンクリックで購読解除できます。アドレスはブリーフィングの送信のみに使用します。