@inproceedings{wang-etal-2019-self,
title = "Self-Attention with Structural Position Representations",
author = "Wang, Xing and
Tu, Zhaopeng and
Wang, Longyue and
Shi, Shuming",
booktitle = "Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)",
month = nov,
year = "2019",
address = "Hong Kong, China",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/D19-1145",
doi = "10.18653/v1/D19-1145",
pages = "1403--1409",
abstract = "Although self-attention networks (SANs) have advanced the state-of-the-art on various NLP tasks, one criticism of SANs is their ability of encoding positions of input words (Shaw et al., 2018). In this work, we propose to augment SANs with structural position representations to model the latent structure of the input sentence, which is complementary to the standard sequential positional representations. Specifically, we use dependency tree to represent the grammatical structure of a sentence, and propose two strategies to encode the positional relationships among words in the dependency tree. Experimental results on NIST Chinese-to-English and WMT14 English-to-German translation tasks show that the proposed approach consistently boosts performance over both the absolute and relative sequential position representations.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="wang-etal-2019-self">
<titleInfo>
<title>Self-Attention with Structural Position Representations</title>
</titleInfo>
<name type="personal">
<namePart type="given">Xing</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zhaopeng</namePart>
<namePart type="family">Tu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Longyue</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Shuming</namePart>
<namePart type="family">Shi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2019-nov</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)</title>
</titleInfo>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Hong Kong, China</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Although self-attention networks (SANs) have advanced the state-of-the-art on various NLP tasks, one criticism of SANs is their ability of encoding positions of input words (Shaw et al., 2018). In this work, we propose to augment SANs with structural position representations to model the latent structure of the input sentence, which is complementary to the standard sequential positional representations. Specifically, we use dependency tree to represent the grammatical structure of a sentence, and propose two strategies to encode the positional relationships among words in the dependency tree. Experimental results on NIST Chinese-to-English and WMT14 English-to-German translation tasks show that the proposed approach consistently boosts performance over both the absolute and relative sequential position representations.</abstract>
<identifier type="citekey">wang-etal-2019-self</identifier>
<identifier type="doi">10.18653/v1/D19-1145</identifier>
<location>
<url>https://aclanthology.org/D19-1145</url>
</location>
<part>
<date>2019-nov</date>
<extent unit="page">
<start>1403</start>
<end>1409</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Self-Attention with Structural Position Representations
%A Wang, Xing
%A Tu, Zhaopeng
%A Wang, Longyue
%A Shi, Shuming
%S Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP)
%D 2019
%8 nov
%I Association for Computational Linguistics
%C Hong Kong, China
%F wang-etal-2019-self
%X Although self-attention networks (SANs) have advanced the state-of-the-art on various NLP tasks, one criticism of SANs is their ability of encoding positions of input words (Shaw et al., 2018). In this work, we propose to augment SANs with structural position representations to model the latent structure of the input sentence, which is complementary to the standard sequential positional representations. Specifically, we use dependency tree to represent the grammatical structure of a sentence, and propose two strategies to encode the positional relationships among words in the dependency tree. Experimental results on NIST Chinese-to-English and WMT14 English-to-German translation tasks show that the proposed approach consistently boosts performance over both the absolute and relative sequential position representations.
%R 10.18653/v1/D19-1145
%U https://aclanthology.org/D19-1145
%U https://doi.org/10.18653/v1/D19-1145
%P 1403-1409
Markdown (Informal)
[Self-Attention with Structural Position Representations](https://aclanthology.org/D19-1145) (Wang et al., EMNLP 2019)
ACL
- Xing Wang, Zhaopeng Tu, Longyue Wang, and Shuming Shi. 2019. Self-Attention with Structural Position Representations. In Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing (EMNLP-IJCNLP), pages 1403–1409, Hong Kong, China. Association for Computational Linguistics.