@inproceedings{nguyen-etal-2024-democratizing,
title = "Democratizing {LLM}s for Low-Resource Languages by Leveraging their {E}nglish Dominant Abilities with Linguistically-Diverse Prompts",
author = "Nguyen, Xuan-Phi and
Aljunied, Mahani and
Joty, Shafiq and
Bing, Lidong",
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.192",
doi = "10.18653/v1/2024.acl-long.192",
pages = "3501--3516",
abstract = "Large language models (LLMs) are known to effectively perform tasks by simply observing few exemplars. However, in low-resource languages, obtaining such hand-picked exemplars can still be challenging, where unsupervised techniques may be necessary. Moreover, competent generative capabilities of LLMs are observed only in high-resource languages, while their performances among under-represented languages fall behind due to pre-training data imbalance. To elicit LLMs{'} ability onto low-resource languages without any supervised data, we propose to assemble synthetic exemplars from a diverse set of high-resource languages to prompt the LLMs to translate from any language into English. These prompts are then used to create intra-lingual exemplars to perform tasks in the target languages. Our unsupervised prompting method performs on par with supervised few-shot learning in LLMs of different sizes for translations between English and 13 Indic and 21 African low-resource languages. We also show that fine-tuning a 7B model on data generated from our method helps it perform competitively with a 175B model. In non-English translation tasks, our method even outperforms supervised prompting by up to 3 chrF++ in many low-resource languages. When evaluated on zero-shot multilingual summarization, our method surpasses other English-pivoting baselines by up to 4 ROUGE-L and is also favored by GPT-4.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="nguyen-etal-2024-democratizing">
<titleInfo>
<title>Democratizing LLMs for Low-Resource Languages by Leveraging their English Dominant Abilities with Linguistically-Diverse Prompts</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xuan-Phi</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mahani</namePart>
<namePart type="family">Aljunied</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shafiq</namePart>
<namePart type="family">Joty</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lidong</namePart>
<namePart type="family">Bing</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) are known to effectively perform tasks by simply observing few exemplars. However, in low-resource languages, obtaining such hand-picked exemplars can still be challenging, where unsupervised techniques may be necessary. Moreover, competent generative capabilities of LLMs are observed only in high-resource languages, while their performances among under-represented languages fall behind due to pre-training data imbalance. To elicit LLMs’ ability onto low-resource languages without any supervised data, we propose to assemble synthetic exemplars from a diverse set of high-resource languages to prompt the LLMs to translate from any language into English. These prompts are then used to create intra-lingual exemplars to perform tasks in the target languages. Our unsupervised prompting method performs on par with supervised few-shot learning in LLMs of different sizes for translations between English and 13 Indic and 21 African low-resource languages. We also show that fine-tuning a 7B model on data generated from our method helps it perform competitively with a 175B model. In non-English translation tasks, our method even outperforms supervised prompting by up to 3 chrF++ in many low-resource languages. When evaluated on zero-shot multilingual summarization, our method surpasses other English-pivoting baselines by up to 4 ROUGE-L and is also favored by GPT-4.</abstract>
<identifier type="citekey">nguyen-etal-2024-democratizing</identifier>
<identifier type="doi">10.18653/v1/2024.acl-long.192</identifier>
<location>
<url>https://aclanthology.org/2024.acl-long.192</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>3501</start>
<end>3516</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Democratizing LLMs for Low-Resource Languages by Leveraging their English Dominant Abilities with Linguistically-Diverse Prompts
%A Nguyen, Xuan-Phi
%A Aljunied, Mahani
%A Joty, Shafiq
%A Bing, Lidong
%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 nguyen-etal-2024-democratizing
%X Large language models (LLMs) are known to effectively perform tasks by simply observing few exemplars. However, in low-resource languages, obtaining such hand-picked exemplars can still be challenging, where unsupervised techniques may be necessary. Moreover, competent generative capabilities of LLMs are observed only in high-resource languages, while their performances among under-represented languages fall behind due to pre-training data imbalance. To elicit LLMs’ ability onto low-resource languages without any supervised data, we propose to assemble synthetic exemplars from a diverse set of high-resource languages to prompt the LLMs to translate from any language into English. These prompts are then used to create intra-lingual exemplars to perform tasks in the target languages. Our unsupervised prompting method performs on par with supervised few-shot learning in LLMs of different sizes for translations between English and 13 Indic and 21 African low-resource languages. We also show that fine-tuning a 7B model on data generated from our method helps it perform competitively with a 175B model. In non-English translation tasks, our method even outperforms supervised prompting by up to 3 chrF++ in many low-resource languages. When evaluated on zero-shot multilingual summarization, our method surpasses other English-pivoting baselines by up to 4 ROUGE-L and is also favored by GPT-4.
%R 10.18653/v1/2024.acl-long.192
%U https://aclanthology.org/2024.acl-long.192
%U https://doi.org/10.18653/v1/2024.acl-long.192
%P 3501-3516
Markdown (Informal)
[Democratizing LLMs for Low-Resource Languages by Leveraging their English Dominant Abilities with Linguistically-Diverse Prompts](https://aclanthology.org/2024.acl-long.192) (Nguyen et al., ACL 2024)
ACL