@inproceedings{rubino-2020-nict,
title = "{NICT} {K}yoto Submission for the {WMT}{'}20 Quality Estimation Task: Intermediate Training for Domain and Task Adaptation",
author = "Rubino, Raphael",
booktitle = "Proceedings of the Fifth Conference on Machine Translation",
month = nov,
year = "2020",
address = "Online",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2020.wmt-1.121",
pages = "1042--1048",
abstract = "This paper describes the NICT Kyoto submission for the WMT{'}20 Quality Estimation (QE) shared task. We participated in Task 2: Word and Sentence-level Post-editing Effort, which involved Wikipedia data and two translation directions, namely English-to-German and English-to-Chinese. Our approach is based on multi-task fine-tuned cross-lingual language models (XLM), initially pre-trained and further domain-adapted through intermediate training using the translation language model (TLM) approach complemented with a novel self-supervised learning task which aim is to model errors inherent to machine translation outputs. Results obtained on both word and sentence-level QE show that the proposed intermediate training method is complementary to language model domain adaptation and outperforms the fine-tuning only approach.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rubino-2020-nict">
<titleInfo>
<title>NICT Kyoto Submission for the WMT’20 Quality Estimation Task: Intermediate Training for Domain and Task Adaptation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Raphael</namePart>
<namePart type="family">Rubino</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-nov</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Fifth Conference on Machine Translation</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>This paper describes the NICT Kyoto submission for the WMT’20 Quality Estimation (QE) shared task. We participated in Task 2: Word and Sentence-level Post-editing Effort, which involved Wikipedia data and two translation directions, namely English-to-German and English-to-Chinese. Our approach is based on multi-task fine-tuned cross-lingual language models (XLM), initially pre-trained and further domain-adapted through intermediate training using the translation language model (TLM) approach complemented with a novel self-supervised learning task which aim is to model errors inherent to machine translation outputs. Results obtained on both word and sentence-level QE show that the proposed intermediate training method is complementary to language model domain adaptation and outperforms the fine-tuning only approach.</abstract>
<identifier type="citekey">rubino-2020-nict</identifier>
<location>
<url>https://aclanthology.org/2020.wmt-1.121</url>
</location>
<part>
<date>2020-nov</date>
<extent unit="page">
<start>1042</start>
<end>1048</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T NICT Kyoto Submission for the WMT’20 Quality Estimation Task: Intermediate Training for Domain and Task Adaptation
%A Rubino, Raphael
%S Proceedings of the Fifth Conference on Machine Translation
%D 2020
%8 nov
%I Association for Computational Linguistics
%C Online
%F rubino-2020-nict
%X This paper describes the NICT Kyoto submission for the WMT’20 Quality Estimation (QE) shared task. We participated in Task 2: Word and Sentence-level Post-editing Effort, which involved Wikipedia data and two translation directions, namely English-to-German and English-to-Chinese. Our approach is based on multi-task fine-tuned cross-lingual language models (XLM), initially pre-trained and further domain-adapted through intermediate training using the translation language model (TLM) approach complemented with a novel self-supervised learning task which aim is to model errors inherent to machine translation outputs. Results obtained on both word and sentence-level QE show that the proposed intermediate training method is complementary to language model domain adaptation and outperforms the fine-tuning only approach.
%U https://aclanthology.org/2020.wmt-1.121
%P 1042-1048
Markdown (Informal)
[NICT Kyoto Submission for the WMT’20 Quality Estimation Task: Intermediate Training for Domain and Task Adaptation](https://aclanthology.org/2020.wmt-1.121) (Rubino, WMT 2020)
ACL