@inproceedings{raganato-etal-2021-empirical,
title = "An Empirical Investigation of Word Alignment Supervision for Zero-Shot Multilingual Neural Machine Translation",
author = {Raganato, Alessandro and
V{\'a}zquez, Ra{\'u}l and
Creutz, Mathias and
Tiedemann, J{\"o}rg},
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-main.664",
doi = "10.18653/v1/2021.emnlp-main.664",
pages = "8449--8456",
abstract = "Zero-shot translations is a fascinating feature of Multilingual Neural Machine Translation (MNMT) systems. These MNMT models are usually trained on English-centric data, i.e. English either as the source or target language, and with a language label prepended to the input indicating the target language. However, recent work has highlighted several flaws of these models in zero-shot scenarios where language labels are ignored and the wrong language is generated or different runs show highly unstable results. In this paper, we investigate the benefits of an explicit alignment to language labels in Transformer-based MNMT models in the zero-shot context, by jointly training one cross attention head with word alignment supervision to stress the focus on the target language label. We compare and evaluate several MNMT systems on three multilingual MT benchmarks of different sizes, showing that simply supervising one cross attention head to focus both on word alignments and language labels reduces the bias towards translating into the wrong language, improving the zero-shot performance overall. Moreover, as an additional advantage, we find that our alignment supervision leads to more stable results across different training runs.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="raganato-etal-2021-empirical">
<titleInfo>
<title>An Empirical Investigation of Word Alignment Supervision for Zero-Shot Multilingual Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Raganato</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Raúl</namePart>
<namePart type="family">Vázquez</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mathias</namePart>
<namePart type="family">Creutz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jörg</namePart>
<namePart type="family">Tiedemann</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marie-Francine</namePart>
<namePart type="family">Moens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanjing</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Specia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="given">Wen-tau</namePart>
<namePart type="family">Yih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online and Punta Cana, Dominican Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Zero-shot translations is a fascinating feature of Multilingual Neural Machine Translation (MNMT) systems. These MNMT models are usually trained on English-centric data, i.e. English either as the source or target language, and with a language label prepended to the input indicating the target language. However, recent work has highlighted several flaws of these models in zero-shot scenarios where language labels are ignored and the wrong language is generated or different runs show highly unstable results. In this paper, we investigate the benefits of an explicit alignment to language labels in Transformer-based MNMT models in the zero-shot context, by jointly training one cross attention head with word alignment supervision to stress the focus on the target language label. We compare and evaluate several MNMT systems on three multilingual MT benchmarks of different sizes, showing that simply supervising one cross attention head to focus both on word alignments and language labels reduces the bias towards translating into the wrong language, improving the zero-shot performance overall. Moreover, as an additional advantage, we find that our alignment supervision leads to more stable results across different training runs.</abstract>
<identifier type="citekey">raganato-etal-2021-empirical</identifier>
<identifier type="doi">10.18653/v1/2021.emnlp-main.664</identifier>
<location>
<url>https://aclanthology.org/2021.emnlp-main.664</url>
</location>
<part>
<date>2021-11</date>
<extent unit="page">
<start>8449</start>
<end>8456</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T An Empirical Investigation of Word Alignment Supervision for Zero-Shot Multilingual Neural Machine Translation
%A Raganato, Alessandro
%A Vázquez, Raúl
%A Creutz, Mathias
%A Tiedemann, Jörg
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
%D 2021
%8 November
%I Association for Computational Linguistics
%C Online and Punta Cana, Dominican Republic
%F raganato-etal-2021-empirical
%X Zero-shot translations is a fascinating feature of Multilingual Neural Machine Translation (MNMT) systems. These MNMT models are usually trained on English-centric data, i.e. English either as the source or target language, and with a language label prepended to the input indicating the target language. However, recent work has highlighted several flaws of these models in zero-shot scenarios where language labels are ignored and the wrong language is generated or different runs show highly unstable results. In this paper, we investigate the benefits of an explicit alignment to language labels in Transformer-based MNMT models in the zero-shot context, by jointly training one cross attention head with word alignment supervision to stress the focus on the target language label. We compare and evaluate several MNMT systems on three multilingual MT benchmarks of different sizes, showing that simply supervising one cross attention head to focus both on word alignments and language labels reduces the bias towards translating into the wrong language, improving the zero-shot performance overall. Moreover, as an additional advantage, we find that our alignment supervision leads to more stable results across different training runs.
%R 10.18653/v1/2021.emnlp-main.664
%U https://aclanthology.org/2021.emnlp-main.664
%U https://doi.org/10.18653/v1/2021.emnlp-main.664
%P 8449-8456
Markdown (Informal)
[An Empirical Investigation of Word Alignment Supervision for Zero-Shot Multilingual Neural Machine Translation](https://aclanthology.org/2021.emnlp-main.664) (Raganato et al., EMNLP 2021)
ACL