@inproceedings{enomoto-etal-2025-fair,
title = "A Fair Comparison without Translationese: {E}nglish vs. Target-language Instructions for Multilingual {LLM}s",
author = "Enomoto, Taisei and
Kim, Hwichan and
Chen, Zhousi and
Komachi, Mamoru",
editor = "Chiruzzo, Luis and
Ritter, Alan and
Wang, Lu",
booktitle = "Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)",
month = apr,
year = "2025",
address = "Albuquerque, New Mexico",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.naacl-short.55/",
doi = "10.18653/v1/2025.naacl-short.55",
pages = "649--670",
ISBN = "979-8-89176-190-2",
abstract = "Most large language models are multilingual instruction executors. Prior studies suggested that English instructions are more effective than target-language instructions even for non-English tasks; however, these studies often use datasets and instructions translated from English, which introduce biases known as translationese, hindering an unbiased comparison. To address this issue, we conduct a fair comparison between English and target-language instructions by eliminating translationese effects. Contrary to previous studies, our experiments across several tasks reveal that the advantage of adopting English instructions is not overwhelming. Additionally, we report on the features of generated texts and the instruction-following abilities when using respective instructions."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="enomoto-etal-2025-fair">
<titleInfo>
<title>A Fair Comparison without Translationese: English vs. Target-language Instructions for Multilingual LLMs</title>
</titleInfo>
<name type="personal">
<namePart type="given">Taisei</namePart>
<namePart type="family">Enomoto</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hwichan</namePart>
<namePart type="family">Kim</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhousi</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mamoru</namePart>
<namePart type="family">Komachi</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>Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)</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">979-8-89176-190-2</identifier>
</relatedItem>
<abstract>Most large language models are multilingual instruction executors. Prior studies suggested that English instructions are more effective than target-language instructions even for non-English tasks; however, these studies often use datasets and instructions translated from English, which introduce biases known as translationese, hindering an unbiased comparison. To address this issue, we conduct a fair comparison between English and target-language instructions by eliminating translationese effects. Contrary to previous studies, our experiments across several tasks reveal that the advantage of adopting English instructions is not overwhelming. Additionally, we report on the features of generated texts and the instruction-following abilities when using respective instructions.</abstract>
<identifier type="citekey">enomoto-etal-2025-fair</identifier>
<identifier type="doi">10.18653/v1/2025.naacl-short.55</identifier>
<location>
<url>https://aclanthology.org/2025.naacl-short.55/</url>
</location>
<part>
<date>2025-04</date>
<extent unit="page">
<start>649</start>
<end>670</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A Fair Comparison without Translationese: English vs. Target-language Instructions for Multilingual LLMs
%A Enomoto, Taisei
%A Kim, Hwichan
%A Chen, Zhousi
%A Komachi, Mamoru
%Y Chiruzzo, Luis
%Y Ritter, Alan
%Y Wang, Lu
%S Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies (Volume 2: Short Papers)
%D 2025
%8 April
%I Association for Computational Linguistics
%C Albuquerque, New Mexico
%@ 979-8-89176-190-2
%F enomoto-etal-2025-fair
%X Most large language models are multilingual instruction executors. Prior studies suggested that English instructions are more effective than target-language instructions even for non-English tasks; however, these studies often use datasets and instructions translated from English, which introduce biases known as translationese, hindering an unbiased comparison. To address this issue, we conduct a fair comparison between English and target-language instructions by eliminating translationese effects. Contrary to previous studies, our experiments across several tasks reveal that the advantage of adopting English instructions is not overwhelming. Additionally, we report on the features of generated texts and the instruction-following abilities when using respective instructions.
%R 10.18653/v1/2025.naacl-short.55
%U https://aclanthology.org/2025.naacl-short.55/
%U https://doi.org/10.18653/v1/2025.naacl-short.55
%P 649-670
Markdown (Informal)
[A Fair Comparison without Translationese: English vs. Target-language Instructions for Multilingual LLMs](https://aclanthology.org/2025.naacl-short.55/) (Enomoto et al., NAACL 2025)
ACL