@inproceedings{cheng-etal-2024-scale,
title = "{SCALE}: Synergized Collaboration of Asymmetric Language Translation Engines",
author = "Cheng, Xin and
Wang, Xun and
Ge, Tao and
Chen, Si-Qing and
Wei, Furu and
Zhao, Dongyan and
Yan, Rui",
editor = "Ku, Lun-Wei and
Martins, Andre and
Srikumar, Vivek",
booktitle = "Findings of the Association for Computational Linguistics ACL 2024",
month = aug,
year = "2024",
address = "Bangkok, Thailand and virtual meeting",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-acl.941",
pages = "15903--15918",
abstract = "In this paper, we introduce SCALE, a collaborative framework that connects a compact Specialized Translation Model (STM) and a general-purpose Large Language Model (LLM) as one unified translation engine. By introducing translation from STM into the triplet in-context demonstrations, SCALE unlocks refinement and pivoting ability of LLM, thus 1) mitigating language bias of LLMs and parallel data bias of STMs, 2) enhancing LLM speciality without sacrificing generality, and 3) facilitating continual learning in a LLM-tuning-free way.Our comprehensive experiments show that SCALE significantly outperforms both LLMs (GPT-4, GPT-3.5) and supervised models (NLLB, M2M) in either high-resource or challenging low-resource settings. Moreover SCALE shows great scalability by only updating the lightweight STM and witness consistent system improvement, an averaged 4 BLEURT score across 4 languages without tuning LLM. Interestingly, SCALE could also effectively exploit the existing language bias of LLMs by using an English-centric STM as a pivot to conduct translation between any language pairs, outperforming GPT-4 by an average of 6 COMET points across eight translation directions. Furthermore we provide an in-depth analysis of SCALE{'}s robustness, translation characteristics, latency costs and inherent language bias, providing solid foundation for future studies exploring the potential synergy between LLMs and more specialized models.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="cheng-etal-2024-scale">
<titleInfo>
<title>SCALE: Synergized Collaboration of Asymmetric Language Translation Engines</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xin</namePart>
<namePart type="family">Cheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xun</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tao</namePart>
<namePart type="family">Ge</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Si-Qing</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Furu</namePart>
<namePart type="family">Wei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dongyan</namePart>
<namePart type="family">Zhao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rui</namePart>
<namePart type="family">Yan</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>Findings of the Association for Computational Linguistics ACL 2024</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 and virtual meeting</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>In this paper, we introduce SCALE, a collaborative framework that connects a compact Specialized Translation Model (STM) and a general-purpose Large Language Model (LLM) as one unified translation engine. By introducing translation from STM into the triplet in-context demonstrations, SCALE unlocks refinement and pivoting ability of LLM, thus 1) mitigating language bias of LLMs and parallel data bias of STMs, 2) enhancing LLM speciality without sacrificing generality, and 3) facilitating continual learning in a LLM-tuning-free way.Our comprehensive experiments show that SCALE significantly outperforms both LLMs (GPT-4, GPT-3.5) and supervised models (NLLB, M2M) in either high-resource or challenging low-resource settings. Moreover SCALE shows great scalability by only updating the lightweight STM and witness consistent system improvement, an averaged 4 BLEURT score across 4 languages without tuning LLM. Interestingly, SCALE could also effectively exploit the existing language bias of LLMs by using an English-centric STM as a pivot to conduct translation between any language pairs, outperforming GPT-4 by an average of 6 COMET points across eight translation directions. Furthermore we provide an in-depth analysis of SCALE’s robustness, translation characteristics, latency costs and inherent language bias, providing solid foundation for future studies exploring the potential synergy between LLMs and more specialized models.</abstract>
<identifier type="citekey">cheng-etal-2024-scale</identifier>
<location>
<url>https://aclanthology.org/2024.findings-acl.941</url>
</location>
<part>
<date>2024-08</date>
<extent unit="page">
<start>15903</start>
<end>15918</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T SCALE: Synergized Collaboration of Asymmetric Language Translation Engines
%A Cheng, Xin
%A Wang, Xun
%A Ge, Tao
%A Chen, Si-Qing
%A Wei, Furu
%A Zhao, Dongyan
%A Yan, Rui
%Y Ku, Lun-Wei
%Y Martins, Andre
%Y Srikumar, Vivek
%S Findings of the Association for Computational Linguistics ACL 2024
%D 2024
%8 August
%I Association for Computational Linguistics
%C Bangkok, Thailand and virtual meeting
%F cheng-etal-2024-scale
%X In this paper, we introduce SCALE, a collaborative framework that connects a compact Specialized Translation Model (STM) and a general-purpose Large Language Model (LLM) as one unified translation engine. By introducing translation from STM into the triplet in-context demonstrations, SCALE unlocks refinement and pivoting ability of LLM, thus 1) mitigating language bias of LLMs and parallel data bias of STMs, 2) enhancing LLM speciality without sacrificing generality, and 3) facilitating continual learning in a LLM-tuning-free way.Our comprehensive experiments show that SCALE significantly outperforms both LLMs (GPT-4, GPT-3.5) and supervised models (NLLB, M2M) in either high-resource or challenging low-resource settings. Moreover SCALE shows great scalability by only updating the lightweight STM and witness consistent system improvement, an averaged 4 BLEURT score across 4 languages without tuning LLM. Interestingly, SCALE could also effectively exploit the existing language bias of LLMs by using an English-centric STM as a pivot to conduct translation between any language pairs, outperforming GPT-4 by an average of 6 COMET points across eight translation directions. Furthermore we provide an in-depth analysis of SCALE’s robustness, translation characteristics, latency costs and inherent language bias, providing solid foundation for future studies exploring the potential synergy between LLMs and more specialized models.
%U https://aclanthology.org/2024.findings-acl.941
%P 15903-15918
Markdown (Informal)
[SCALE: Synergized Collaboration of Asymmetric Language Translation Engines](https://aclanthology.org/2024.findings-acl.941) (Cheng et al., Findings 2024)
ACL