@inproceedings{sun-etal-2022-alleviating,
title = "Alleviating the Inequality of Attention Heads for Neural Machine Translation",
author = "Sun, Zewei and
Huang, Shujian and
Dai, Xinyu and
Chen, Jiajun",
booktitle = "Proceedings of the 29th International Conference on Computational Linguistics",
month = oct,
year = "2022",
address = "Gyeongju, Republic of Korea",
publisher = "International Committee on Computational Linguistics",
url = "https://aclanthology.org/2022.coling-1.466",
pages = "5246--5250",
abstract = "Recent studies show that the attention heads in Transformer are not equal. We relate this phenomenon to the imbalance training of multi-head attention and the model dependence on specific heads. To tackle this problem, we propose a simple masking method: HeadMask, in two specific ways. Experiments show that translation improvements are achieved on multiple language pairs. Subsequent empirical analyses also support our assumption and confirm the effectiveness of the method.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sun-etal-2022-alleviating">
<titleInfo>
<title>Alleviating the Inequality of Attention Heads for Neural Machine Translation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zewei</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shujian</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xinyu</namePart>
<namePart type="family">Dai</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jiajun</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-10</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 29th International Conference on Computational Linguistics</title>
</titleInfo>
<originInfo>
<publisher>International Committee on Computational Linguistics</publisher>
<place>
<placeTerm type="text">Gyeongju, Republic of Korea</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recent studies show that the attention heads in Transformer are not equal. We relate this phenomenon to the imbalance training of multi-head attention and the model dependence on specific heads. To tackle this problem, we propose a simple masking method: HeadMask, in two specific ways. Experiments show that translation improvements are achieved on multiple language pairs. Subsequent empirical analyses also support our assumption and confirm the effectiveness of the method.</abstract>
<identifier type="citekey">sun-etal-2022-alleviating</identifier>
<location>
<url>https://aclanthology.org/2022.coling-1.466</url>
</location>
<part>
<date>2022-10</date>
<extent unit="page">
<start>5246</start>
<end>5250</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Alleviating the Inequality of Attention Heads for Neural Machine Translation
%A Sun, Zewei
%A Huang, Shujian
%A Dai, Xinyu
%A Chen, Jiajun
%S Proceedings of the 29th International Conference on Computational Linguistics
%D 2022
%8 October
%I International Committee on Computational Linguistics
%C Gyeongju, Republic of Korea
%F sun-etal-2022-alleviating
%X Recent studies show that the attention heads in Transformer are not equal. We relate this phenomenon to the imbalance training of multi-head attention and the model dependence on specific heads. To tackle this problem, we propose a simple masking method: HeadMask, in two specific ways. Experiments show that translation improvements are achieved on multiple language pairs. Subsequent empirical analyses also support our assumption and confirm the effectiveness of the method.
%U https://aclanthology.org/2022.coling-1.466
%P 5246-5250
Markdown (Informal)
[Alleviating the Inequality of Attention Heads for Neural Machine Translation](https://aclanthology.org/2022.coling-1.466) (Sun et al., COLING 2022)
ACL