@inproceedings{xuewen-etal-2021-reducing,
title = "Reducing Length Bias in Scoring Neural Machine Translation via a Causal Inference Method",
author = "Xuewen, Shi and
Heyan, Huang and
Ping, Jian and
Yi-Kun, Tang",
editor = "Li, Sheng and
Sun, Maosong and
Liu, Yang and
Wu, Hua and
Liu, Kang and
Che, Wanxiang and
He, Shizhu and
Rao, Gaoqi",
booktitle = "Proceedings of the 20th Chinese National Conference on Computational Linguistics",
month = aug,
year = "2021",
address = "Huhhot, China",
publisher = "Chinese Information Processing Society of China",
url = "https://aclanthology.org/2021.ccl-1.78/",
pages = "874--885",
language = "eng",
abstract = "Neural machine translation (NMT) usually employs beam search to expand the searching spaceand obtain more translation candidates. However the increase of the beam size often suffersfrom plenty of short translations resulting in dramatical decrease in translation quality. In this paper we handle the length bias problem through a perspective of causal inference. Specially we regard the model generated translation score S as a degraded true translation quality affectedby some noise and one of the confounders is the translation length. We apply a Half-Sibling Re-gression method to remove the length effect on S and then we can obtain a debiased translation score without length information. The proposed method is model agnostic and unsupervised which is adaptive to any NMT model and test dataset. We conduct the experiments on three translation tasks with different scales of datasets. Experimental results and further analyses showthat our approaches gain comparable performance with the empirical baseline methods."
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xuewen-etal-2021-reducing">
<titleInfo>
<title>Reducing Length Bias in Scoring Neural Machine Translation via a Causal Inference Method</title>
</titleInfo>
<name type="personal">
<namePart type="given">Shi</namePart>
<namePart type="family">Xuewen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Huang</namePart>
<namePart type="family">Heyan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jian</namePart>
<namePart type="family">Ping</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Tang</namePart>
<namePart type="family">Yi-Kun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-08</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<language>
<languageTerm type="text">eng</languageTerm>
</language>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 20th Chinese National Conference on Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sheng</namePart>
<namePart type="family">Li</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<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">Hua</namePart>
<namePart type="family">Wu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kang</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">Shizhu</namePart>
<namePart type="family">He</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>
<originInfo>
<publisher>Chinese Information Processing Society of China</publisher>
<place>
<placeTerm type="text">Huhhot, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Neural machine translation (NMT) usually employs beam search to expand the searching spaceand obtain more translation candidates. However the increase of the beam size often suffersfrom plenty of short translations resulting in dramatical decrease in translation quality. In this paper we handle the length bias problem through a perspective of causal inference. Specially we regard the model generated translation score S as a degraded true translation quality affectedby some noise and one of the confounders is the translation length. We apply a Half-Sibling Re-gression method to remove the length effect on S and then we can obtain a debiased translation score without length information. The proposed method is model agnostic and unsupervised which is adaptive to any NMT model and test dataset. We conduct the experiments on three translation tasks with different scales of datasets. Experimental results and further analyses showthat our approaches gain comparable performance with the empirical baseline methods.</abstract>
<identifier type="citekey">xuewen-etal-2021-reducing</identifier>
<location>
<url>https://aclanthology.org/2021.ccl-1.78/</url>
</location>
<part>
<date>2021-08</date>
<extent unit="page">
<start>874</start>
<end>885</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Reducing Length Bias in Scoring Neural Machine Translation via a Causal Inference Method
%A Xuewen, Shi
%A Heyan, Huang
%A Ping, Jian
%A Yi-Kun, Tang
%Y Li, Sheng
%Y Sun, Maosong
%Y Liu, Yang
%Y Wu, Hua
%Y Liu, Kang
%Y Che, Wanxiang
%Y He, Shizhu
%Y Rao, Gaoqi
%S Proceedings of the 20th Chinese National Conference on Computational Linguistics
%D 2021
%8 August
%I Chinese Information Processing Society of China
%C Huhhot, China
%G eng
%F xuewen-etal-2021-reducing
%X Neural machine translation (NMT) usually employs beam search to expand the searching spaceand obtain more translation candidates. However the increase of the beam size often suffersfrom plenty of short translations resulting in dramatical decrease in translation quality. In this paper we handle the length bias problem through a perspective of causal inference. Specially we regard the model generated translation score S as a degraded true translation quality affectedby some noise and one of the confounders is the translation length. We apply a Half-Sibling Re-gression method to remove the length effect on S and then we can obtain a debiased translation score without length information. The proposed method is model agnostic and unsupervised which is adaptive to any NMT model and test dataset. We conduct the experiments on three translation tasks with different scales of datasets. Experimental results and further analyses showthat our approaches gain comparable performance with the empirical baseline methods.
%U https://aclanthology.org/2021.ccl-1.78/
%P 874-885
Markdown (Informal)
[Reducing Length Bias in Scoring Neural Machine Translation via a Causal Inference Method](https://aclanthology.org/2021.ccl-1.78/) (Xuewen et al., CCL 2021)
ACL