@inproceedings{poncelas-etal-2018-investigating,
title = "Investigating Backtranslation in Neural Machine Translation",
author = "Poncelas, Alberto and
Shterionov, Dimitar and
Way, Andy and
Maillette de Buy Wenniger, Gideon and
Passban, Peyman",
editor = "P{\'e}rez-Ortiz, Juan Antonio and
S{\'a}nchez-Mart{\'\i}nez, Felipe and
Espl{\`a}-Gomis, Miquel and
Popovi{\'c}, Maja and
Rico, Celia and
Martins, Andr{\'e} and
Van den Bogaert, Joachim and
Forcada, Mikel L.",
booktitle = "Proceedings of the 21st Annual Conference of the European Association for Machine Translation",
month = may,
year = "2018",
address = "Alicante, Spain",
url = "https://aclanthology.org/2018.eamt-main.25",
pages = "269--278",
abstract = "A prerequisite for training corpus-based machine translation (MT) systems {--} either Statistical MT (SMT) or Neural MT (NMT) {--} is the availability of high-quality parallel data. This is arguably more important today than ever before, as NMT has been shown in many studies to outperform SMT, but mostly when large parallel corpora are available; in cases where data is limited, SMT can still outperform NMT. Recently researchers have shown that back-translating monolingual data can be used to create synthetic parallel corpora, which in turn can be used in combination with authentic parallel data to train a highquality NMT system. Given that large collections of new parallel text become available only quite rarely, backtranslation has become the norm when building state-of-the-art NMT systems, especially in resource-poor scenarios. However, we assert that there are many unknown factors regarding the actual effects of back-translated data on the translation capabilities of an NMT model. Accordingly, in this work we investigate how using back-translated data as a training corpus {--} both as a separate standalone dataset as well as combined with human-generated parallel data {--} affects the performance of an NMT model. We use incrementally larger amounts of back-translated data to train a range of NMT systems for German-to-English, and analyse the resulting translation performance.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="poncelas-etal-2018-investigating">
<titleInfo>
<title>Investigating Backtranslation in Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Alberto</namePart>
<namePart type="family">Poncelas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Dimitar</namePart>
<namePart type="family">Shterionov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Andy</namePart>
<namePart type="family">Way</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gideon</namePart>
<namePart type="family">Maillette de Buy Wenniger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peyman</namePart>
<namePart type="family">Passban</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2018-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 21st Annual Conference of the European Association for Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Juan</namePart>
<namePart type="given">Antonio</namePart>
<namePart type="family">Pérez-Ortiz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Felipe</namePart>
<namePart type="family">Sánchez-Martínez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Miquel</namePart>
<namePart type="family">Esplà-Gomis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maja</namePart>
<namePart type="family">Popović</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Celia</namePart>
<namePart type="family">Rico</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">André</namePart>
<namePart type="family">Martins</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joachim</namePart>
<namePart type="family">Van den Bogaert</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mikel</namePart>
<namePart type="given">L</namePart>
<namePart type="family">Forcada</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<place>
<placeTerm type="text">Alicante, Spain</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>A prerequisite for training corpus-based machine translation (MT) systems – either Statistical MT (SMT) or Neural MT (NMT) – is the availability of high-quality parallel data. This is arguably more important today than ever before, as NMT has been shown in many studies to outperform SMT, but mostly when large parallel corpora are available; in cases where data is limited, SMT can still outperform NMT. Recently researchers have shown that back-translating monolingual data can be used to create synthetic parallel corpora, which in turn can be used in combination with authentic parallel data to train a highquality NMT system. Given that large collections of new parallel text become available only quite rarely, backtranslation has become the norm when building state-of-the-art NMT systems, especially in resource-poor scenarios. However, we assert that there are many unknown factors regarding the actual effects of back-translated data on the translation capabilities of an NMT model. Accordingly, in this work we investigate how using back-translated data as a training corpus – both as a separate standalone dataset as well as combined with human-generated parallel data – affects the performance of an NMT model. We use incrementally larger amounts of back-translated data to train a range of NMT systems for German-to-English, and analyse the resulting translation performance.</abstract>
<identifier type="citekey">poncelas-etal-2018-investigating</identifier>
<location>
<url>https://aclanthology.org/2018.eamt-main.25</url>
</location>
<part>
<date>2018-05</date>
<extent unit="page">
<start>269</start>
<end>278</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Investigating Backtranslation in Neural Machine Translation
%A Poncelas, Alberto
%A Shterionov, Dimitar
%A Way, Andy
%A Maillette de Buy Wenniger, Gideon
%A Passban, Peyman
%Y Pérez-Ortiz, Juan Antonio
%Y Sánchez-Martínez, Felipe
%Y Esplà-Gomis, Miquel
%Y Popović, Maja
%Y Rico, Celia
%Y Martins, André
%Y Van den Bogaert, Joachim
%Y Forcada, Mikel L.
%S Proceedings of the 21st Annual Conference of the European Association for Machine Translation
%D 2018
%8 May
%C Alicante, Spain
%F poncelas-etal-2018-investigating
%X A prerequisite for training corpus-based machine translation (MT) systems – either Statistical MT (SMT) or Neural MT (NMT) – is the availability of high-quality parallel data. This is arguably more important today than ever before, as NMT has been shown in many studies to outperform SMT, but mostly when large parallel corpora are available; in cases where data is limited, SMT can still outperform NMT. Recently researchers have shown that back-translating monolingual data can be used to create synthetic parallel corpora, which in turn can be used in combination with authentic parallel data to train a highquality NMT system. Given that large collections of new parallel text become available only quite rarely, backtranslation has become the norm when building state-of-the-art NMT systems, especially in resource-poor scenarios. However, we assert that there are many unknown factors regarding the actual effects of back-translated data on the translation capabilities of an NMT model. Accordingly, in this work we investigate how using back-translated data as a training corpus – both as a separate standalone dataset as well as combined with human-generated parallel data – affects the performance of an NMT model. We use incrementally larger amounts of back-translated data to train a range of NMT systems for German-to-English, and analyse the resulting translation performance.
%U https://aclanthology.org/2018.eamt-main.25
%P 269-278
Markdown (Informal)
[Investigating Backtranslation in Neural Machine Translation](https://aclanthology.org/2018.eamt-main.25) (Poncelas et al., EAMT 2018)
ACL
- Alberto Poncelas, Dimitar Shterionov, Andy Way, Gideon Maillette de Buy Wenniger, and Peyman Passban. 2018. Investigating Backtranslation in Neural Machine Translation. In Proceedings of the 21st Annual Conference of the European Association for Machine Translation, pages 269–278, Alicante, Spain.