@inproceedings{lai-nissim-2024-mcot,
title = "m{C}o{T}: Multilingual Instruction Tuning for Reasoning Consistency in Language Models",
author = "Lai, Huiyuan and
Nissim, Malvina",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = aug,
year = "2024",
address = "Bangkok, Thailand",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.acl-long.649",
doi = "10.18653/v1/2024.acl-long.649",
pages = "12012--12026",
abstract = "Large language models (LLMs) with Chain-of-thought (CoT) have recently emerged as a powerful technique for eliciting reasoning to improve various downstream tasks. As most research mainly focuses on English, with few explorations in a multilingual context, the question of how reliable this reasoning capability is in different languages is still open. To address it directly, we study multilingual reasoning consistency across multiple languages, using popular open-source LLMs. First, we compile the first large-scale multilingual math reasoning dataset, *mCoT-MATH*, covering eleven diverse languages. Then, we introduce multilingual CoT instruction tuning to boost reasoning capability across languages, thereby improving model consistency. While existing LLMs show substantial variation across the languages we consider, and especially low performance for lesser resourced languages, our 7B parameter model *mCoT* achieves impressive consistency across languages, and superior or comparable performance to close- and open-source models even of much larger sizes.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lai-nissim-2024-mcot">
<titleInfo>
<title>mCoT: Multilingual Instruction Tuning for Reasoning Consistency in Language Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Huiyuan</namePart>
<namePart type="family">Lai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Malvina</namePart>
<namePart type="family">Nissim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Lun-Wei</namePart>
<namePart type="family">Ku</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andre</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Vivek</namePart>
<namePart type="family">Srikumar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Bangkok, Thailand</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Large language models (LLMs) with Chain-of-thought (CoT) have recently emerged as a powerful technique for eliciting reasoning to improve various downstream tasks. As most research mainly focuses on English, with few explorations in a multilingual context, the question of how reliable this reasoning capability is in different languages is still open. To address it directly, we study multilingual reasoning consistency across multiple languages, using popular open-source LLMs. First, we compile the first large-scale multilingual math reasoning dataset, *mCoT-MATH*, covering eleven diverse languages. Then, we introduce multilingual CoT instruction tuning to boost reasoning capability across languages, thereby improving model consistency. While existing LLMs show substantial variation across the languages we consider, and especially low performance for lesser resourced languages, our 7B parameter model *mCoT* achieves impressive consistency across languages, and superior or comparable performance to close- and open-source models even of much larger sizes.</abstract>
<identifier type="citekey">lai-nissim-2024-mcot</identifier>
<identifier type="doi">10.18653/v1/2024.acl-long.649</identifier>
<location>
<url>https://aclanthology.org/2024.acl-long.649</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>12012</start>
<end>12026</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T mCoT: Multilingual Instruction Tuning for Reasoning Consistency in Language Models
%A Lai, Huiyuan
%A Nissim, Malvina
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand
%F lai-nissim-2024-mcot
%X Large language models (LLMs) with Chain-of-thought (CoT) have recently emerged as a powerful technique for eliciting reasoning to improve various downstream tasks. As most research mainly focuses on English, with few explorations in a multilingual context, the question of how reliable this reasoning capability is in different languages is still open. To address it directly, we study multilingual reasoning consistency across multiple languages, using popular open-source LLMs. First, we compile the first large-scale multilingual math reasoning dataset, *mCoT-MATH*, covering eleven diverse languages. Then, we introduce multilingual CoT instruction tuning to boost reasoning capability across languages, thereby improving model consistency. While existing LLMs show substantial variation across the languages we consider, and especially low performance for lesser resourced languages, our 7B parameter model *mCoT* achieves impressive consistency across languages, and superior or comparable performance to close- and open-source models even of much larger sizes.
%R 10.18653/v1/2024.acl-long.649
%U https://aclanthology.org/2024.acl-long.649
%U https://doi.org/10.18653/v1/2024.acl-long.649
%P 12012-12026
Markdown (Informal)
[mCoT: Multilingual Instruction Tuning for Reasoning Consistency in Language Models](https://aclanthology.org/2024.acl-long.649) (Lai & Nissim, ACL 2024)
ACL