MyCulture: Exploring Malaysia's Diverse Culture under Low-Resource Language Constraints

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Hew, Zhong Ken, Low, Jia Xin, Yang, Sze Jue, Chan, Chee Seng
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913979713978368
author Hew, Zhong Ken
Low, Jia Xin
Yang, Sze Jue
Chan, Chee Seng
author_facet Hew, Zhong Ken
Low, Jia Xin
Yang, Sze Jue
Chan, Chee Seng
contents Large Language Models (LLMs) often exhibit cultural biases due to training data dominated by high-resource languages like English and Chinese. This poses challenges for accurately representing and evaluating diverse cultural contexts, particularly in low-resource language settings. To address this, we introduce MyCulture, a benchmark designed to comprehensively evaluate LLMs on Malaysian culture across six pillars: arts, attire, customs, entertainment, food, and religion presented in Bahasa Melayu. Unlike conventional benchmarks, MyCulture employs a novel open-ended multiple-choice question format without predefined options, thereby reducing guessing and mitigating format bias. We provide a theoretical justification for the effectiveness of this open-ended structure in improving both fairness and discriminative power. Furthermore, we analyze structural bias by comparing model performance on structured versus free-form outputs, and assess language bias through multilingual prompt variations. Our evaluation across a range of regional and international LLMs reveals significant disparities in cultural comprehension, highlighting the urgent need for culturally grounded and linguistically inclusive benchmarks in the development and assessment of LLMs.
format Preprint
id arxiv_https___arxiv_org_abs_2508_05429
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle MyCulture: Exploring Malaysia's Diverse Culture under Low-Resource Language Constraints
Hew, Zhong Ken
Low, Jia Xin
Yang, Sze Jue
Chan, Chee Seng
Computation and Language
Artificial Intelligence
Large Language Models (LLMs) often exhibit cultural biases due to training data dominated by high-resource languages like English and Chinese. This poses challenges for accurately representing and evaluating diverse cultural contexts, particularly in low-resource language settings. To address this, we introduce MyCulture, a benchmark designed to comprehensively evaluate LLMs on Malaysian culture across six pillars: arts, attire, customs, entertainment, food, and religion presented in Bahasa Melayu. Unlike conventional benchmarks, MyCulture employs a novel open-ended multiple-choice question format without predefined options, thereby reducing guessing and mitigating format bias. We provide a theoretical justification for the effectiveness of this open-ended structure in improving both fairness and discriminative power. Furthermore, we analyze structural bias by comparing model performance on structured versus free-form outputs, and assess language bias through multilingual prompt variations. Our evaluation across a range of regional and international LLMs reveals significant disparities in cultural comprehension, highlighting the urgent need for culturally grounded and linguistically inclusive benchmarks in the development and assessment of LLMs.
title MyCulture: Exploring Malaysia's Diverse Culture under Low-Resource Language Constraints
topic Computation and Language
Artificial Intelligence
url https://arxiv.org/abs/2508.05429