@inproceedings{wu-etal-2025-combining,
title = "Combining the Best of Both Worlds: A Method for Hybrid {NMT} and {LLM} Translation",
author = "Wu, Zhanglin and
Wei, Daimeng and
Chen, Xiaoyu and
Shang, Hengchao and
Guo, Jiaxin and
Li, Zongyao and
Luo, Yuanchang and
Yang, Jinlong and
Rao, Zhiqiang and
Yang, Hao",
editor = "Che, Wanxiang and
Nabende, Joyce and
Shutova, Ekaterina and
Pilehvar, Mohammad Taher",
booktitle = "Findings of the Association for Computational Linguistics: ACL 2025",
month = jul,
year = "2025",
address = "Vienna, Austria",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2025.findings-acl.266/",
doi = "10.18653/v1/2025.findings-acl.266",
pages = "5140--5148",
ISBN = "979-8-89176-256-5",
abstract = "Large language model (LLM) shows promising performances in a variety of downstream tasks, such as machine translation (MT). However, using LLMs for translation suffers from high computational costs and significant latency. Based on our evaluation, in most cases, translations using LLMs are comparable to that generated by neural machine translation (NMT) systems. Only in particular scenarios, LLM and NMT models show respective advantages. As a result, integrating NMT and LLM for translation and using LLM only when necessary seems to be a sound solution. A scheduling policy that optimizes translation result while ensuring fast speed and as less LLM usage as possible is thereby required. We compare several scheduling policies and propose a novel and straightforward decider that leverages source sentence features. We conduct extensive experiments on multilingual test sets and the result shows that we can achieve optimal translation performance with less LLM usage, demonstrating effectiveness of our decider."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wu-etal-2025-combining">
<titleInfo>
<title>Combining the Best of Both Worlds: A Method for Hybrid NMT and LLM Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhanglin</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Daimeng</namePart>
<namePart type="family">Wei</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xiaoyu</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hengchao</namePart>
<namePart type="family">Shang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiaxin</namePart>
<namePart type="family">Guo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zongyao</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuanchang</namePart>
<namePart type="family">Luo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jinlong</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhiqiang</namePart>
<namePart type="family">Rao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hao</namePart>
<namePart type="family">Yang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2025-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: ACL 2025</title>
</titleInfo>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joyce</namePart>
<namePart type="family">Nabende</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ekaterina</namePart>
<namePart type="family">Shutova</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Vienna, Austria</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-8-89176-256-5</identifier>
</relatedItem>
<abstract>Large language model (LLM) shows promising performances in a variety of downstream tasks, such as machine translation (MT). However, using LLMs for translation suffers from high computational costs and significant latency. Based on our evaluation, in most cases, translations using LLMs are comparable to that generated by neural machine translation (NMT) systems. Only in particular scenarios, LLM and NMT models show respective advantages. As a result, integrating NMT and LLM for translation and using LLM only when necessary seems to be a sound solution. A scheduling policy that optimizes translation result while ensuring fast speed and as less LLM usage as possible is thereby required. We compare several scheduling policies and propose a novel and straightforward decider that leverages source sentence features. We conduct extensive experiments on multilingual test sets and the result shows that we can achieve optimal translation performance with less LLM usage, demonstrating effectiveness of our decider.</abstract>
<identifier type="citekey">wu-etal-2025-combining</identifier>
<identifier type="doi">10.18653/v1/2025.findings-acl.266</identifier>
<location>
<url>https://aclanthology.org/2025.findings-acl.266/</url>
</location>
<part>
<date>2025-07</date>
<extent unit="page">
<start>5140</start>
<end>5148</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Combining the Best of Both Worlds: A Method for Hybrid NMT and LLM Translation
%A Wu, Zhanglin
%A Wei, Daimeng
%A Chen, Xiaoyu
%A Shang, Hengchao
%A Guo, Jiaxin
%A Li, Zongyao
%A Luo, Yuanchang
%A Yang, Jinlong
%A Rao, Zhiqiang
%A Yang, Hao
%Y Che, Wanxiang
%Y Nabende, Joyce
%Y Shutova, Ekaterina
%Y Pilehvar, Mohammad Taher
%S Findings of the Association for Computational Linguistics: ACL 2025
%D 2025
%8 July
%I Association for Computational Linguistics
%C Vienna, Austria
%@ 979-8-89176-256-5
%F wu-etal-2025-combining
%X Large language model (LLM) shows promising performances in a variety of downstream tasks, such as machine translation (MT). However, using LLMs for translation suffers from high computational costs and significant latency. Based on our evaluation, in most cases, translations using LLMs are comparable to that generated by neural machine translation (NMT) systems. Only in particular scenarios, LLM and NMT models show respective advantages. As a result, integrating NMT and LLM for translation and using LLM only when necessary seems to be a sound solution. A scheduling policy that optimizes translation result while ensuring fast speed and as less LLM usage as possible is thereby required. We compare several scheduling policies and propose a novel and straightforward decider that leverages source sentence features. We conduct extensive experiments on multilingual test sets and the result shows that we can achieve optimal translation performance with less LLM usage, demonstrating effectiveness of our decider.
%R 10.18653/v1/2025.findings-acl.266
%U https://aclanthology.org/2025.findings-acl.266/
%U https://doi.org/10.18653/v1/2025.findings-acl.266
%P 5140-5148
Markdown (Informal)
[Combining the Best of Both Worlds: A Method for Hybrid NMT and LLM Translation](https://aclanthology.org/2025.findings-acl.266/) (Wu et al., Findings 2025)
ACL
- Zhanglin Wu, Daimeng Wei, Xiaoyu Chen, Hengchao Shang, Jiaxin Guo, Zongyao Li, Yuanchang Luo, Jinlong Yang, Zhiqiang Rao, and Hao Yang. 2025. Combining the Best of Both Worlds: A Method for Hybrid NMT and LLM Translation. In Findings of the Association for Computational Linguistics: ACL 2025, pages 5140–5148, Vienna, Austria. Association for Computational Linguistics.