@inproceedings{li-etal-2022-universal,
title = "Universal Conditional Masked Language Pre-training for Neural Machine Translation",
author = "Li, Pengfei and
Li, Liangyou and
Zhang, Meng and
Wu, Minghao and
Liu, Qun",
editor = "Muresan, Smaranda and
Nakov, Preslav and
Villavicencio, Aline",
booktitle = "Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = may,
year = "2022",
address = "Dublin, Ireland",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.acl-long.442",
doi = "10.18653/v1/2022.acl-long.442",
pages = "6379--6391",
abstract = "Pre-trained sequence-to-sequence models have significantly improved Neural Machine Translation (NMT). Different from prior works where pre-trained models usually adopt an unidirectional decoder, this paper demonstrates that pre-training a sequence-to-sequence model but with a bidirectional decoder can produce notable performance gains for both Autoregressive and Non-autoregressive NMT. Specifically, we propose CeMAT, a conditional masked language model pre-trained on large-scale bilingual and monolingual corpora in many languages. We also introduce two simple but effective methods to enhance the CeMAT, aligned code-switching {\&} masking and dynamic dual-masking. We conduct extensive experiments and show that our CeMAT can achieve significant performance improvement for all scenarios from low- to extremely high-resource languages, i.e., up to +14.4 BLEU on low resource and +7.9 BLEU improvements on average for Autoregressive NMT. For Non-autoregressive NMT, we demonstrate it can also produce consistent performance gains, i.e., up to +5.3 BLEU. To the best of our knowledge, this is the first work to pre-train a unified model for fine-tuning on both NMT tasks. Code, data, and pre-trained models are available at \url{https://github.com/huawei-noah/Pretrained-Language-Model/CeMAT}",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="li-etal-2022-universal">
<titleInfo>
<title>Universal Conditional Masked Language Pre-training for Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Pengfei</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liangyou</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Meng</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Minghao</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Qun</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Smaranda</namePart>
<namePart type="family">Muresan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Preslav</namePart>
<namePart type="family">Nakov</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aline</namePart>
<namePart type="family">Villavicencio</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dublin, Ireland</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Pre-trained sequence-to-sequence models have significantly improved Neural Machine Translation (NMT). Different from prior works where pre-trained models usually adopt an unidirectional decoder, this paper demonstrates that pre-training a sequence-to-sequence model but with a bidirectional decoder can produce notable performance gains for both Autoregressive and Non-autoregressive NMT. Specifically, we propose CeMAT, a conditional masked language model pre-trained on large-scale bilingual and monolingual corpora in many languages. We also introduce two simple but effective methods to enhance the CeMAT, aligned code-switching & masking and dynamic dual-masking. We conduct extensive experiments and show that our CeMAT can achieve significant performance improvement for all scenarios from low- to extremely high-resource languages, i.e., up to +14.4 BLEU on low resource and +7.9 BLEU improvements on average for Autoregressive NMT. For Non-autoregressive NMT, we demonstrate it can also produce consistent performance gains, i.e., up to +5.3 BLEU. To the best of our knowledge, this is the first work to pre-train a unified model for fine-tuning on both NMT tasks. Code, data, and pre-trained models are available at https://github.com/huawei-noah/Pretrained-Language-Model/CeMAT</abstract>
<identifier type="citekey">li-etal-2022-universal</identifier>
<identifier type="doi">10.18653/v1/2022.acl-long.442</identifier>
<location>
<url>https://aclanthology.org/2022.acl-long.442</url>
</location>
<part>
<date>2022-05</date>
<extent unit="page">
<start>6379</start>
<end>6391</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Universal Conditional Masked Language Pre-training for Neural Machine Translation
%A Li, Pengfei
%A Li, Liangyou
%A Zhang, Meng
%A Wu, Minghao
%A Liu, Qun
%Y Muresan, Smaranda
%Y Nakov, Preslav
%Y Villavicencio, Aline
%S Proceedings of the 60th Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2022
%8 May
%I Association for Computational Linguistics
%C Dublin, Ireland
%F li-etal-2022-universal
%X Pre-trained sequence-to-sequence models have significantly improved Neural Machine Translation (NMT). Different from prior works where pre-trained models usually adopt an unidirectional decoder, this paper demonstrates that pre-training a sequence-to-sequence model but with a bidirectional decoder can produce notable performance gains for both Autoregressive and Non-autoregressive NMT. Specifically, we propose CeMAT, a conditional masked language model pre-trained on large-scale bilingual and monolingual corpora in many languages. We also introduce two simple but effective methods to enhance the CeMAT, aligned code-switching & masking and dynamic dual-masking. We conduct extensive experiments and show that our CeMAT can achieve significant performance improvement for all scenarios from low- to extremely high-resource languages, i.e., up to +14.4 BLEU on low resource and +7.9 BLEU improvements on average for Autoregressive NMT. For Non-autoregressive NMT, we demonstrate it can also produce consistent performance gains, i.e., up to +5.3 BLEU. To the best of our knowledge, this is the first work to pre-train a unified model for fine-tuning on both NMT tasks. Code, data, and pre-trained models are available at https://github.com/huawei-noah/Pretrained-Language-Model/CeMAT
%R 10.18653/v1/2022.acl-long.442
%U https://aclanthology.org/2022.acl-long.442
%U https://doi.org/10.18653/v1/2022.acl-long.442
%P 6379-6391
Markdown (Informal)
[Universal Conditional Masked Language Pre-training for Neural Machine Translation](https://aclanthology.org/2022.acl-long.442) (Li et al., ACL 2022)
ACL