Measuring Hong Kong Massive Multi-Task Language Understanding

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Cao, Chuxue, Zhu, Zhenghao, Zhu, Junqi, Lu, Guoying, Peng, Siyu, Dai, Juntao, Shi, Weijie, Han, Sirui, Guo, Yike
Format: Preprint
Published: 2025
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866915273059074048
author Cao, Chuxue
Zhu, Zhenghao
Zhu, Junqi
Lu, Guoying
Peng, Siyu
Dai, Juntao
Shi, Weijie
Han, Sirui
Guo, Yike
author_facet Cao, Chuxue
Zhu, Zhenghao
Zhu, Junqi
Lu, Guoying
Peng, Siyu
Dai, Juntao
Shi, Weijie
Han, Sirui
Guo, Yike
contents Multilingual understanding is crucial for the cross-cultural applicability of Large Language Models (LLMs). However, evaluation benchmarks designed for Hong Kong's unique linguistic landscape, which combines Traditional Chinese script with Cantonese as the spoken form and its cultural context, remain underdeveloped. To address this gap, we introduce HKMMLU, a multi-task language understanding benchmark that evaluates Hong Kong's linguistic competence and socio-cultural knowledge. The HKMMLU includes 26,698 multi-choice questions across 66 subjects, organized into four categories: Science, Technology, Engineering, and Mathematics (STEM), Social Sciences, Humanities, and Other. To evaluate the multilingual understanding ability of LLMs, 90,550 Mandarin-Cantonese translation tasks were additionally included. We conduct comprehensive experiments on GPT-4o, Claude 3.7 Sonnet, and 18 open-source LLMs of varying sizes on HKMMLU. The results show that the best-performing model, DeepSeek-V3, struggles to achieve an accuracy of 75\%, significantly lower than that of MMLU and CMMLU. This performance gap highlights the need to improve LLMs' capabilities in Hong Kong-specific language and knowledge domains. Furthermore, we investigate how question language, model size, prompting strategies, and question and reasoning token lengths affect model performance. We anticipate that HKMMLU will significantly advance the development of LLMs in multilingual and cross-cultural contexts, thereby enabling broader and more impactful applications.
format Preprint
id arxiv_https___arxiv_org_abs_2505_02177
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Measuring Hong Kong Massive Multi-Task Language Understanding
Cao, Chuxue
Zhu, Zhenghao
Zhu, Junqi
Lu, Guoying
Peng, Siyu
Dai, Juntao
Shi, Weijie
Han, Sirui
Guo, Yike
Computation and Language
Multilingual understanding is crucial for the cross-cultural applicability of Large Language Models (LLMs). However, evaluation benchmarks designed for Hong Kong's unique linguistic landscape, which combines Traditional Chinese script with Cantonese as the spoken form and its cultural context, remain underdeveloped. To address this gap, we introduce HKMMLU, a multi-task language understanding benchmark that evaluates Hong Kong's linguistic competence and socio-cultural knowledge. The HKMMLU includes 26,698 multi-choice questions across 66 subjects, organized into four categories: Science, Technology, Engineering, and Mathematics (STEM), Social Sciences, Humanities, and Other. To evaluate the multilingual understanding ability of LLMs, 90,550 Mandarin-Cantonese translation tasks were additionally included. We conduct comprehensive experiments on GPT-4o, Claude 3.7 Sonnet, and 18 open-source LLMs of varying sizes on HKMMLU. The results show that the best-performing model, DeepSeek-V3, struggles to achieve an accuracy of 75\%, significantly lower than that of MMLU and CMMLU. This performance gap highlights the need to improve LLMs' capabilities in Hong Kong-specific language and knowledge domains. Furthermore, we investigate how question language, model size, prompting strategies, and question and reasoning token lengths affect model performance. We anticipate that HKMMLU will significantly advance the development of LLMs in multilingual and cross-cultural contexts, thereby enabling broader and more impactful applications.
title Measuring Hong Kong Massive Multi-Task Language Understanding
topic Computation and Language
url https://arxiv.org/abs/2505.02177