Saved in:
Bibliographic Details
Main Authors: Zhou, Huichi, Xu, Zehao, Zhao, Munan, Li, Kaihong, Li, Yiqiang, Wang, Hongtao
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2504.19759
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866916710064324608
author Zhou, Huichi
Xu, Zehao
Zhao, Munan
Li, Kaihong
Li, Yiqiang
Wang, Hongtao
author_facet Zhou, Huichi
Xu, Zehao
Zhao, Munan
Li, Kaihong
Li, Yiqiang
Wang, Hongtao
contents In this paper, we introduce the Multilingual Moral Reasoning Benchmark (MMRB) to evaluate the moral reasoning abilities of large language models (LLMs) across five typologically diverse languages and three levels of contextual complexity: sentence, paragraph, and document. Our results show moral reasoning performance degrades with increasing context complexity, particularly for low-resource languages such as Vietnamese. We further fine-tune the open-source LLaMA-3-8B model using curated monolingual data for alignment and poisoning. Surprisingly, low-resource languages have a stronger impact on multilingual reasoning than high-resource ones, highlighting their critical role in multilingual NLP.
format Preprint
id arxiv_https___arxiv_org_abs_2504_19759
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Moral Reasoning Across Languages: The Critical Role of Low-Resource Languages in LLMs
Zhou, Huichi
Xu, Zehao
Zhao, Munan
Li, Kaihong
Li, Yiqiang
Wang, Hongtao
Computation and Language
In this paper, we introduce the Multilingual Moral Reasoning Benchmark (MMRB) to evaluate the moral reasoning abilities of large language models (LLMs) across five typologically diverse languages and three levels of contextual complexity: sentence, paragraph, and document. Our results show moral reasoning performance degrades with increasing context complexity, particularly for low-resource languages such as Vietnamese. We further fine-tune the open-source LLaMA-3-8B model using curated monolingual data for alignment and poisoning. Surprisingly, low-resource languages have a stronger impact on multilingual reasoning than high-resource ones, highlighting their critical role in multilingual NLP.
title Moral Reasoning Across Languages: The Critical Role of Low-Resource Languages in LLMs
topic Computation and Language
url https://arxiv.org/abs/2504.19759