@inproceedings{wang-etal-2021-niutrans,
title = "The {N}iu{T}rans System for the {WMT} 2021 Efficiency Task",
author = "Wang, Chenglong and
Hu, Chi and
Mu, Yongyu and
Yan, Zhongxiang and
Wu, Siming and
Hu, Yimin and
Cao, Hang and
Li, Bei and
Lin, Ye and
Xiao, Tong and
Zhu, Jingbo",
booktitle = "Proceedings of the Sixth Conference on Machine Translation",
month = nov,
year = "2021",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.wmt-1.76",
pages = "787--794",
abstract = "This paper describes the NiuTrans system for the WMT21 translation efficiency task. Following last year{'}s work, we explore various techniques to improve the efficiency while maintaining translation quality. We investigate the combinations of lightweight Transformer architectures and knowledge distillation strategies. Also, we improve the translation efficiency with graph optimization, low precision, dynamic batching, and parallel pre/post-processing. Putting these together, our system can translate 247,000 words per second on an NVIDIA A100, being 3$\times$ faster than our last year{'}s system. Our system is the fastest and has the lowest memory consumption on the GPU-throughput track. The code, model, and pipeline will be available at NiuTrans.NMT.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2021-niutrans">
<titleInfo>
<title>The NiuTrans System for the WMT 2021 Efficiency Task</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chenglong</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chi</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yongyu</namePart>
<namePart type="family">Mu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhongxiang</namePart>
<namePart type="family">Yan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Siming</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yimin</namePart>
<namePart type="family">Hu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hang</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bei</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ye</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tong</namePart>
<namePart type="family">Xiao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jingbo</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-nov</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Sixth Conference on Machine Translation</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper describes the NiuTrans system for the WMT21 translation efficiency task. Following last year’s work, we explore various techniques to improve the efficiency while maintaining translation quality. We investigate the combinations of lightweight Transformer architectures and knowledge distillation strategies. Also, we improve the translation efficiency with graph optimization, low precision, dynamic batching, and parallel pre/post-processing. Putting these together, our system can translate 247,000 words per second on an NVIDIA A100, being 3$\times$ faster than our last year’s system. Our system is the fastest and has the lowest memory consumption on the GPU-throughput track. The code, model, and pipeline will be available at NiuTrans.NMT.</abstract>
<identifier type="citekey">wang-etal-2021-niutrans</identifier>
<location>
<url>https://aclanthology.org/2021.wmt-1.76</url>
</location>
<part>
<date>2021-nov</date>
<extent unit="page">
<start>787</start>
<end>794</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The NiuTrans System for the WMT 2021 Efficiency Task
%A Wang, Chenglong
%A Hu, Chi
%A Mu, Yongyu
%A Yan, Zhongxiang
%A Wu, Siming
%A Hu, Yimin
%A Cao, Hang
%A Li, Bei
%A Lin, Ye
%A Xiao, Tong
%A Zhu, Jingbo
%S Proceedings of the Sixth Conference on Machine Translation
%D 2021
%8 nov
%I Association for Computational Linguistics
%C Online
%F wang-etal-2021-niutrans
%X This paper describes the NiuTrans system for the WMT21 translation efficiency task. Following last year’s work, we explore various techniques to improve the efficiency while maintaining translation quality. We investigate the combinations of lightweight Transformer architectures and knowledge distillation strategies. Also, we improve the translation efficiency with graph optimization, low precision, dynamic batching, and parallel pre/post-processing. Putting these together, our system can translate 247,000 words per second on an NVIDIA A100, being 3$\times$ faster than our last year’s system. Our system is the fastest and has the lowest memory consumption on the GPU-throughput track. The code, model, and pipeline will be available at NiuTrans.NMT.
%U https://aclanthology.org/2021.wmt-1.76
%P 787-794
Markdown (Informal)
[The NiuTrans System for the WMT 2021 Efficiency Task](https://aclanthology.org/2021.wmt-1.76) (Wang et al., WMT 2021)
ACL
- Chenglong Wang, Chi Hu, Yongyu Mu, Zhongxiang Yan, Siming Wu, Yimin Hu, Hang Cao, Bei Li, Ye Lin, Tong Xiao, and Jingbo Zhu. 2021. The NiuTrans System for the WMT 2021 Efficiency Task. In Proceedings of the Sixth Conference on Machine Translation, pages 787–794, Online. Association for Computational Linguistics.