@inproceedings{yang-etal-2023-local,
title = "Local Interpretation of Transformer Based on Linear Decomposition",
author = "Yang, Sen and
Huang, Shujian and
Zou, Wei and
Zhang, Jianbing and
Dai, Xinyu and
Chen, Jiajun",
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.572",
doi = "10.18653/v1/2023.acl-long.572",
pages = "10270--10287",
abstract = "In recent years, deep neural networks (DNNs) have achieved state-of-the-art performance on a wide range of tasks. However, limitations in interpretability have hindered their applications in the real world. This work proposes to interpret neural networks by linear decomposition and finds that the ReLU-activated Transformer can be considered as a linear model on a single input. We further leverage the linearity of the model and propose a linear decomposition of the model output to generate local explanations. Our evaluation of sentiment classification and machine translation shows that our method achieves competitive performance in efficiency and fidelity of explanation. In addition, we demonstrate the potential of our approach in applications with examples of error analysis on multiple tasks.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="yang-etal-2023-local">
<titleInfo>
<title>Local Interpretation of Transformer Based on Linear Decomposition</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sen</namePart>
<namePart type="family">Yang</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">Wei</namePart>
<namePart type="family">Zou</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jianbing</namePart>
<namePart type="family">Zhang</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>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>In recent years, deep neural networks (DNNs) have achieved state-of-the-art performance on a wide range of tasks. However, limitations in interpretability have hindered their applications in the real world. This work proposes to interpret neural networks by linear decomposition and finds that the ReLU-activated Transformer can be considered as a linear model on a single input. We further leverage the linearity of the model and propose a linear decomposition of the model output to generate local explanations. Our evaluation of sentiment classification and machine translation shows that our method achieves competitive performance in efficiency and fidelity of explanation. In addition, we demonstrate the potential of our approach in applications with examples of error analysis on multiple tasks.</abstract>
<identifier type="citekey">yang-etal-2023-local</identifier>
<identifier type="doi">10.18653/v1/2023.acl-long.572</identifier>
<location>
<url>https://aclanthology.org/2023.acl-long.572</url>
</location>
<part>
<date>2023-07</date>
<extent unit="page">
<start>10270</start>
<end>10287</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Local Interpretation of Transformer Based on Linear Decomposition
%A Yang, Sen
%A Huang, Shujian
%A Zou, Wei
%A Zhang, Jianbing
%A Dai, Xinyu
%A Chen, Jiajun
%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 yang-etal-2023-local
%X In recent years, deep neural networks (DNNs) have achieved state-of-the-art performance on a wide range of tasks. However, limitations in interpretability have hindered their applications in the real world. This work proposes to interpret neural networks by linear decomposition and finds that the ReLU-activated Transformer can be considered as a linear model on a single input. We further leverage the linearity of the model and propose a linear decomposition of the model output to generate local explanations. Our evaluation of sentiment classification and machine translation shows that our method achieves competitive performance in efficiency and fidelity of explanation. In addition, we demonstrate the potential of our approach in applications with examples of error analysis on multiple tasks.
%R 10.18653/v1/2023.acl-long.572
%U https://aclanthology.org/2023.acl-long.572
%U https://doi.org/10.18653/v1/2023.acl-long.572
%P 10270-10287
Markdown (Informal)
[Local Interpretation of Transformer Based on Linear Decomposition](https://aclanthology.org/2023.acl-long.572) (Yang et al., ACL 2023)
ACL
- Sen Yang, Shujian Huang, Wei Zou, Jianbing Zhang, Xinyu Dai, and Jiajun Chen. 2023. Local Interpretation of Transformer Based on Linear Decomposition. In Proceedings of the 61st Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers), pages 10270–10287, Toronto, Canada. Association for Computational Linguistics.