@inproceedings{chunyou-etal-2022-towards,
title = "Towards Making the Most of Pre-trained Translation Model for Quality Estimation",
author = "Chunyou, Li and
Hui, Di and
Hui, Huang and
Kazushige, Ouchi and
Yufeng, Chen and
Jian, Liu and
Jinan, Xu",
editor = "Sun, Maosong and
Liu, Yang and
Che, Wanxiang and
Feng, Yang and
Qiu, Xipeng and
Rao, Gaoqi and
Chen, Yubo",
booktitle = "Proceedings of the 21st Chinese National Conference on Computational Linguistics",
month = oct,
year = "2022",
address = "Nanchang, China",
publisher = "Chinese Information Processing Society of China",
url = "https://aclanthology.org/2022.ccl-1.77",
pages = "872--883",
abstract = "{``}Machine translation quality estimation (QE) aims to evaluate the quality of machine translation automatically without relying on any reference. One common practice is applying the translation model as a feature extractor. However, there exist several discrepancies between the translation model and the QE model. The translation model is trained in an autoregressive manner, while the QE model is performed in a non-autoregressive manner. Besides, the translation model only learns to model human-crafted parallel data, while the QE model needs to model machinetranslated noisy data. In order to bridge these discrepancies, we propose two strategies to posttrain the translation model, namely Conditional Masked Language Modeling (CMLM) and Denoising Restoration (DR). Specifically, CMLM learns to predict masked tokens at the target side conditioned on the source sentence. DR firstly introduces noise to the target side of parallel data, and the model is trained to detect and recover the introduced noise. Both strategies can adapt the pre-trained translation model to the QE-style prediction task. Experimental results show that our model achieves impressive results, significantly outperforming the baseline model, verifying the effectiveness of our proposed methods.{''}",
language = "English",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chunyou-etal-2022-towards">
<titleInfo>
<title>Towards Making the Most of Pre-trained Translation Model for Quality Estimation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Li</namePart>
<namePart type="family">Chunyou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Di</namePart>
<namePart type="family">Hui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Huang</namePart>
<namePart type="family">Hui</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ouchi</namePart>
<namePart type="family">Kazushige</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chen</namePart>
<namePart type="family">Yufeng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liu</namePart>
<namePart type="family">Jian</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xu</namePart>
<namePart type="family">Jinan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<language>
<languageTerm type="text">English</languageTerm>
<languageTerm type="code" authority="iso639-2b">eng</languageTerm>
</language>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 21st Chinese National Conference on Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Maosong</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Wanxiang</namePart>
<namePart type="family">Che</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yang</namePart>
<namePart type="family">Feng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xipeng</namePart>
<namePart type="family">Qiu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Gaoqi</namePart>
<namePart type="family">Rao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yubo</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Chinese Information Processing Society of China</publisher>
<place>
<placeTerm type="text">Nanchang, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>“Machine translation quality estimation (QE) aims to evaluate the quality of machine translation automatically without relying on any reference. One common practice is applying the translation model as a feature extractor. However, there exist several discrepancies between the translation model and the QE model. The translation model is trained in an autoregressive manner, while the QE model is performed in a non-autoregressive manner. Besides, the translation model only learns to model human-crafted parallel data, while the QE model needs to model machinetranslated noisy data. In order to bridge these discrepancies, we propose two strategies to posttrain the translation model, namely Conditional Masked Language Modeling (CMLM) and Denoising Restoration (DR). Specifically, CMLM learns to predict masked tokens at the target side conditioned on the source sentence. DR firstly introduces noise to the target side of parallel data, and the model is trained to detect and recover the introduced noise. Both strategies can adapt the pre-trained translation model to the QE-style prediction task. Experimental results show that our model achieves impressive results, significantly outperforming the baseline model, verifying the effectiveness of our proposed methods.”</abstract>
<identifier type="citekey">chunyou-etal-2022-towards</identifier>
<location>
<url>https://aclanthology.org/2022.ccl-1.77</url>
</location>
<part>
<date>2022-10</date>
<extent unit="page">
<start>872</start>
<end>883</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Towards Making the Most of Pre-trained Translation Model for Quality Estimation
%A Chunyou, Li
%A Hui, Di
%A Hui, Huang
%A Kazushige, Ouchi
%A Yufeng, Chen
%A Jian, Liu
%A Jinan, Xu
%Y Sun, Maosong
%Y Liu, Yang
%Y Che, Wanxiang
%Y Feng, Yang
%Y Qiu, Xipeng
%Y Rao, Gaoqi
%Y Chen, Yubo
%S Proceedings of the 21st Chinese National Conference on Computational Linguistics
%D 2022
%8 October
%I Chinese Information Processing Society of China
%C Nanchang, China
%G English
%F chunyou-etal-2022-towards
%X “Machine translation quality estimation (QE) aims to evaluate the quality of machine translation automatically without relying on any reference. One common practice is applying the translation model as a feature extractor. However, there exist several discrepancies between the translation model and the QE model. The translation model is trained in an autoregressive manner, while the QE model is performed in a non-autoregressive manner. Besides, the translation model only learns to model human-crafted parallel data, while the QE model needs to model machinetranslated noisy data. In order to bridge these discrepancies, we propose two strategies to posttrain the translation model, namely Conditional Masked Language Modeling (CMLM) and Denoising Restoration (DR). Specifically, CMLM learns to predict masked tokens at the target side conditioned on the source sentence. DR firstly introduces noise to the target side of parallel data, and the model is trained to detect and recover the introduced noise. Both strategies can adapt the pre-trained translation model to the QE-style prediction task. Experimental results show that our model achieves impressive results, significantly outperforming the baseline model, verifying the effectiveness of our proposed methods.”
%U https://aclanthology.org/2022.ccl-1.77
%P 872-883
Markdown (Informal)
[Towards Making the Most of Pre-trained Translation Model for Quality Estimation](https://aclanthology.org/2022.ccl-1.77) (Chunyou et al., CCL 2022)
ACL