@inproceedings{ngo-etal-2020-improving,
title = "Improving Multilingual Neural Machine Translation For Low-Resource Languages: {F}rench, {E}nglish - {V}ietnamese",
author = "Ngo, Thi-Vinh and
Nguyen, Phuong-Thai and
Ha, Thanh-Le and
Dinh, Khac-Quy and
Nguyen, Le-Minh",
booktitle = "Proceedings of the 3rd Workshop on Technologies for MT of Low Resource Languages",
month = dec,
year = "2020",
address = "Suzhou, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.loresmt-1.8",
pages = "55--61",
abstract = "Prior works have demonstrated that a low-resource language pair can benefit from multilingual machine translation (MT) systems, which rely on many language pairs{'} joint training. This paper proposes two simple strategies to address the rare word issue in multilingual MT systems for two low-resource language pairs: French-Vietnamese and English-Vietnamese. The first strategy is about dynamical learning word similarity of tokens in the shared space among source languages while another one attempts to augment the translation ability of rare words through updating their embeddings during the training. Besides, we leverage monolingual data for multilingual MT systems to increase the amount of synthetic parallel corpora while dealing with the data sparsity problem. We have shown significant improvements of up to +1.62 and +2.54 BLEU points over the bilingual baseline systems for both language pairs and released our datasets for the research community.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="ngo-etal-2020-improving">
<titleInfo>
<title>Improving Multilingual Neural Machine Translation For Low-Resource Languages: French, English - Vietnamese</title>
</titleInfo>
<name type="personal">
<namePart type="given">Thi-Vinh</namePart>
<namePart type="family">Ngo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Phuong-Thai</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thanh-Le</namePart>
<namePart type="family">Ha</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khac-Quy</namePart>
<namePart type="family">Dinh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Le-Minh</namePart>
<namePart type="family">Nguyen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-dec</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 3rd Workshop on Technologies for MT of Low Resource Languages</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Suzhou, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Prior works have demonstrated that a low-resource language pair can benefit from multilingual machine translation (MT) systems, which rely on many language pairs’ joint training. This paper proposes two simple strategies to address the rare word issue in multilingual MT systems for two low-resource language pairs: French-Vietnamese and English-Vietnamese. The first strategy is about dynamical learning word similarity of tokens in the shared space among source languages while another one attempts to augment the translation ability of rare words through updating their embeddings during the training. Besides, we leverage monolingual data for multilingual MT systems to increase the amount of synthetic parallel corpora while dealing with the data sparsity problem. We have shown significant improvements of up to +1.62 and +2.54 BLEU points over the bilingual baseline systems for both language pairs and released our datasets for the research community.</abstract>
<identifier type="citekey">ngo-etal-2020-improving</identifier>
<location>
<url>https://aclanthology.org/2020.loresmt-1.8</url>
</location>
<part>
<date>2020-dec</date>
<extent unit="page">
<start>55</start>
<end>61</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improving Multilingual Neural Machine Translation For Low-Resource Languages: French, English - Vietnamese
%A Ngo, Thi-Vinh
%A Nguyen, Phuong-Thai
%A Ha, Thanh-Le
%A Dinh, Khac-Quy
%A Nguyen, Le-Minh
%S Proceedings of the 3rd Workshop on Technologies for MT of Low Resource Languages
%D 2020
%8 dec
%I Association for Computational Linguistics
%C Suzhou, China
%F ngo-etal-2020-improving
%X Prior works have demonstrated that a low-resource language pair can benefit from multilingual machine translation (MT) systems, which rely on many language pairs’ joint training. This paper proposes two simple strategies to address the rare word issue in multilingual MT systems for two low-resource language pairs: French-Vietnamese and English-Vietnamese. The first strategy is about dynamical learning word similarity of tokens in the shared space among source languages while another one attempts to augment the translation ability of rare words through updating their embeddings during the training. Besides, we leverage monolingual data for multilingual MT systems to increase the amount of synthetic parallel corpora while dealing with the data sparsity problem. We have shown significant improvements of up to +1.62 and +2.54 BLEU points over the bilingual baseline systems for both language pairs and released our datasets for the research community.
%U https://aclanthology.org/2020.loresmt-1.8
%P 55-61
Markdown (Informal)
[Improving Multilingual Neural Machine Translation For Low-Resource Languages: French, English - Vietnamese](https://aclanthology.org/2020.loresmt-1.8) (Ngo et al., loresmt 2020)
ACL