@inproceedings{ranaldi-etal-2025-natural,
title = "When natural language is not enough: The limits of in-context learning demonstrations in multilingual reasoning",
author = "Ranaldi, Leonardo and
Haddow, Barry and
Birch, Alexandra",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Findings of the Association for Computational Linguistics: NAACL 2025",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-naacl.412/",
doi = "10.18653/v1/2025.findings-naacl.412",
pages = "7369--7396",
ISBN = "979-8-89176-195-7",
abstract = "Previous studies have demonstrated the effectiveness of reasoning methods in eliciting multi-step reasoned answers from Large Language Models (LLMs) by leveraging in-context demonstrations. These methods, exemplified by Chain-of-Thought (CoT) and Program-Aided Language Models (PAL), have been shown to perform well in monolingual contexts, primarily in English. There has, however, been limited exploration of their abilities in other languages.To gain a deeper understanding of the role of reasoning methods for in-context demonstrations, we investigate how well CoT and PAL perform across languages for arithmetic and symbolic reasoning tasks. Our findings indicate that the effectiveness of reasoning methods varies significantly across different languages and models. Specifically, CoT, which relies on natural language demonstrations, tends to be more accurate in high-resource than in low-resource languages. Conversely, the structured nature of PAL demonstrations facilitates multilingual comprehension, enabling LLMs to generate programmatic answers in both high- and low-resource languages and leading to significant performance improvements over CoT concerning the accuracy of the generated responses."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ranaldi-etal-2025-natural">
<titleInfo>
<title>When natural language is not enough: The limits of in-context learning demonstrations in multilingual reasoning</title>
</titleInfo>
<name type="personal">
<namePart type="given">Leonardo</namePart>
<namePart type="family">Ranaldi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Barry</namePart>
<namePart type="family">Haddow</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alexandra</namePart>
<namePart type="family">Birch</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-04</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: NAACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Luis</namePart>
<namePart type="family">Chiruzzo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alan</namePart>
<namePart type="family">Ritter</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Albuquerque, New Mexico</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-195-7</identifier>
</relatedItem>
<abstract>Previous studies have demonstrated the effectiveness of reasoning methods in eliciting multi-step reasoned answers from Large Language Models (LLMs) by leveraging in-context demonstrations. These methods, exemplified by Chain-of-Thought (CoT) and Program-Aided Language Models (PAL), have been shown to perform well in monolingual contexts, primarily in English. There has, however, been limited exploration of their abilities in other languages.To gain a deeper understanding of the role of reasoning methods for in-context demonstrations, we investigate how well CoT and PAL perform across languages for arithmetic and symbolic reasoning tasks. Our findings indicate that the effectiveness of reasoning methods varies significantly across different languages and models. Specifically, CoT, which relies on natural language demonstrations, tends to be more accurate in high-resource than in low-resource languages. Conversely, the structured nature of PAL demonstrations facilitates multilingual comprehension, enabling LLMs to generate programmatic answers in both high- and low-resource languages and leading to significant performance improvements over CoT concerning the accuracy of the generated responses.</abstract>
<identifier type="citekey">ranaldi-etal-2025-natural</identifier>
<identifier type="doi">10.18653/v1/2025.findings-naacl.412</identifier>
<location>
<url>https://aclanthology.org/2025.findings-naacl.412/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>7369</start>
<end>7396</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T When natural language is not enough: The limits of in-context learning demonstrations in multilingual reasoning
%A Ranaldi, Leonardo
%A Haddow, Barry
%A Birch, Alexandra
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Findings of the Association for Computational Linguistics: NAACL 2025
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-195-7
%F ranaldi-etal-2025-natural
%X Previous studies have demonstrated the effectiveness of reasoning methods in eliciting multi-step reasoned answers from Large Language Models (LLMs) by leveraging in-context demonstrations. These methods, exemplified by Chain-of-Thought (CoT) and Program-Aided Language Models (PAL), have been shown to perform well in monolingual contexts, primarily in English. There has, however, been limited exploration of their abilities in other languages.To gain a deeper understanding of the role of reasoning methods for in-context demonstrations, we investigate how well CoT and PAL perform across languages for arithmetic and symbolic reasoning tasks. Our findings indicate that the effectiveness of reasoning methods varies significantly across different languages and models. Specifically, CoT, which relies on natural language demonstrations, tends to be more accurate in high-resource than in low-resource languages. Conversely, the structured nature of PAL demonstrations facilitates multilingual comprehension, enabling LLMs to generate programmatic answers in both high- and low-resource languages and leading to significant performance improvements over CoT concerning the accuracy of the generated responses.
%R 10.18653/v1/2025.findings-naacl.412
%U https://aclanthology.org/2025.findings-naacl.412/
%U https://doi.org/10.18653/v1/2025.findings-naacl.412
%P 7369-7396
Markdown (Informal)
[When natural language is not enough: The limits of in-context learning demonstrations in multilingual reasoning](https://aclanthology.org/2025.findings-naacl.412/) (Ranaldi et al., Findings 2025)
ACL