@inproceedings{lakew-etal-2021-zero,
title = "Zero-Shot Neural Machine Translation with Self-Learning Cycle",
author = "Lakew, Surafel M. and
Negri, Matteo and
Turchi, Marco",
editor = "Ortega, John and
Ojha, Atul Kr. and
Kann, Katharina and
Liu, Chao-Hong",
booktitle = "Proceedings of the 4th Workshop on Technologies for MT of Low Resource Languages (LoResMT2021)",
month = aug,
year = "2021",
address = "Virtual",
publisher = "Association for Machine Translation in the Americas",
url = "https://aclanthology.org/2021.mtsummit-loresmt.10",
pages = "96--113",
abstract = "Neural Machine Translation (NMT) approaches employing monolingual data are showing steady improvements in resource-rich conditions. However, evaluations using real-world lowresource languages still result in unsatisfactory performance. This work proposes a novel zeroshot NMT modeling approach that learns without the now-standard assumption of a pivot language sharing parallel data with the zero-shot source and target languages. Our approach is based on three stages: initialization from any pre-trained NMT model observing at least the target language, augmentation of source sides leveraging target monolingual data, and learning to optimize the initial model to the zero-shot pair, where the latter two constitute a selflearning cycle. Empirical findings involving four diverse (in terms of a language family, script and relatedness) zero-shot pairs show the effectiveness of our approach with up to +5.93 BLEU improvement against a supervised bilingual baseline. Compared to unsupervised NMT, consistent improvements are observed even in a domain-mismatch setting, attesting to the usability of our method.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="lakew-etal-2021-zero">
<titleInfo>
<title>Zero-Shot Neural Machine Translation with Self-Learning Cycle</title>
</titleInfo>
<name type="personal">
<namePart type="given">Surafel</namePart>
<namePart type="given">M</namePart>
<namePart type="family">Lakew</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matteo</namePart>
<namePart type="family">Negri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marco</namePart>
<namePart type="family">Turchi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 4th Workshop on Technologies for MT of Low Resource Languages (LoResMT2021)</title>
</titleInfo>
<name type="personal">
<namePart type="given">John</namePart>
<namePart type="family">Ortega</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Atul</namePart>
<namePart type="given">Kr.</namePart>
<namePart type="family">Ojha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Katharina</namePart>
<namePart type="family">Kann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chao-Hong</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Machine Translation in the Americas</publisher>
<place>
<placeTerm type="text">Virtual</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Neural Machine Translation (NMT) approaches employing monolingual data are showing steady improvements in resource-rich conditions. However, evaluations using real-world lowresource languages still result in unsatisfactory performance. This work proposes a novel zeroshot NMT modeling approach that learns without the now-standard assumption of a pivot language sharing parallel data with the zero-shot source and target languages. Our approach is based on three stages: initialization from any pre-trained NMT model observing at least the target language, augmentation of source sides leveraging target monolingual data, and learning to optimize the initial model to the zero-shot pair, where the latter two constitute a selflearning cycle. Empirical findings involving four diverse (in terms of a language family, script and relatedness) zero-shot pairs show the effectiveness of our approach with up to +5.93 BLEU improvement against a supervised bilingual baseline. Compared to unsupervised NMT, consistent improvements are observed even in a domain-mismatch setting, attesting to the usability of our method.</abstract>
<identifier type="citekey">lakew-etal-2021-zero</identifier>
<location>
<url>https://aclanthology.org/2021.mtsummit-loresmt.10</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>96</start>
<end>113</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Zero-Shot Neural Machine Translation with Self-Learning Cycle
%A Lakew, Surafel M.
%A Negri, Matteo
%A Turchi, Marco
%Y Ortega, John
%Y Ojha, Atul Kr.
%Y Kann, Katharina
%Y Liu, Chao-Hong
%S Proceedings of the 4th Workshop on Technologies for MT of Low Resource Languages (LoResMT2021)
%D 2021
%8 August
%I Association for Machine Translation in the Americas
%C Virtual
%F lakew-etal-2021-zero
%X Neural Machine Translation (NMT) approaches employing monolingual data are showing steady improvements in resource-rich conditions. However, evaluations using real-world lowresource languages still result in unsatisfactory performance. This work proposes a novel zeroshot NMT modeling approach that learns without the now-standard assumption of a pivot language sharing parallel data with the zero-shot source and target languages. Our approach is based on three stages: initialization from any pre-trained NMT model observing at least the target language, augmentation of source sides leveraging target monolingual data, and learning to optimize the initial model to the zero-shot pair, where the latter two constitute a selflearning cycle. Empirical findings involving four diverse (in terms of a language family, script and relatedness) zero-shot pairs show the effectiveness of our approach with up to +5.93 BLEU improvement against a supervised bilingual baseline. Compared to unsupervised NMT, consistent improvements are observed even in a domain-mismatch setting, attesting to the usability of our method.
%U https://aclanthology.org/2021.mtsummit-loresmt.10
%P 96-113
Markdown (Informal)
[Zero-Shot Neural Machine Translation with Self-Learning Cycle](https://aclanthology.org/2021.mtsummit-loresmt.10) (Lakew et al., LoResMT 2021)
ACL