@inproceedings{xilong-etal-2023-interpreting,
title = "Interpreting Positional Information in Perspective of Word Order",
author = "Xilong, Zhang and
Ruochen, Liu and
Jin, Liu and
Xuefeng, Liang",
editor = "Rogers, Anna and
Boyd-Graber, Jordan and
Okazaki, Naoaki",
booktitle = "Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)",
month = jul,
year = "2023",
address = "Toronto, Canada",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.acl-long.534",
doi = "10.18653/v1/2023.acl-long.534",
pages = "9600--9613",
abstract = "The attention mechanism is a powerful and effective method utilized in natural language processing. However, it has been observed that this method is insensitive to positional information. Although several studies have attempted to improve positional encoding and investigate the influence of word order perturbation, it remains unclear how positional encoding impacts NLP models from the perspective of word order. In this paper, we aim to shed light on this problem by analyzing the working mechanism of the attention module and investigating the root cause of its inability to encode positional information. Our hypothesis is that the insensitivity can be attributed to the weight sum operation utilized in the attention module. To verify this hypothesis, we propose a novel weight concatenation operation and evaluate its efficacy in neural machine translation tasks. Our enhanced experimental results not only reveal that the proposed operation can effectively encode positional information but also confirm our hypothesis.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="xilong-etal-2023-interpreting">
<titleInfo>
<title>Interpreting Positional Information in Perspective of Word Order</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhang</namePart>
<namePart type="family">Xilong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liu</namePart>
<namePart type="family">Ruochen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liu</namePart>
<namePart type="family">Jin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liang</namePart>
<namePart type="family">Xuefeng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Anna</namePart>
<namePart type="family">Rogers</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jordan</namePart>
<namePart type="family">Boyd-Graber</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Naoaki</namePart>
<namePart type="family">Okazaki</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Toronto, Canada</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>The attention mechanism is a powerful and effective method utilized in natural language processing. However, it has been observed that this method is insensitive to positional information. Although several studies have attempted to improve positional encoding and investigate the influence of word order perturbation, it remains unclear how positional encoding impacts NLP models from the perspective of word order. In this paper, we aim to shed light on this problem by analyzing the working mechanism of the attention module and investigating the root cause of its inability to encode positional information. Our hypothesis is that the insensitivity can be attributed to the weight sum operation utilized in the attention module. To verify this hypothesis, we propose a novel weight concatenation operation and evaluate its efficacy in neural machine translation tasks. Our enhanced experimental results not only reveal that the proposed operation can effectively encode positional information but also confirm our hypothesis.</abstract>
<identifier type="citekey">xilong-etal-2023-interpreting</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.534</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.534</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>9600</start>
<end>9613</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Interpreting Positional Information in Perspective of Word Order
%A Xilong, Zhang
%A Ruochen, Liu
%A Jin, Liu
%A Xuefeng, Liang
%Y Rogers, Anna
%Y Boyd-Graber, Jordan
%Y Okazaki, Naoaki
%S Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)
%D 2023
%8 July
%I Association for Computational Linguistics
%C Toronto, Canada
%F xilong-etal-2023-interpreting
%X The attention mechanism is a powerful and effective method utilized in natural language processing. However, it has been observed that this method is insensitive to positional information. Although several studies have attempted to improve positional encoding and investigate the influence of word order perturbation, it remains unclear how positional encoding impacts NLP models from the perspective of word order. In this paper, we aim to shed light on this problem by analyzing the working mechanism of the attention module and investigating the root cause of its inability to encode positional information. Our hypothesis is that the insensitivity can be attributed to the weight sum operation utilized in the attention module. To verify this hypothesis, we propose a novel weight concatenation operation and evaluate its efficacy in neural machine translation tasks. Our enhanced experimental results not only reveal that the proposed operation can effectively encode positional information but also confirm our hypothesis.
%R 10.18653/v1/2023.acl-long.534
%U https://aclanthology.org/2023.acl-long.534
%U https://doi.org/10.18653/v1/2023.acl-long.534
%P 9600-9613
Markdown (Informal)
[Interpreting Positional Information in Perspective of Word Order](https://aclanthology.org/2023.acl-long.534) (Xilong et al., ACL 2023)
ACL