@inproceedings{liu-etal-2022-improve,
title = "Improve Interpretability of Neural Networks via Sparse Contrastive Coding",
author = "Liu, Junhong and
Lin, Yijie and
Jiang, Liang and
Liu, Jia and
Wen, Zujie and
Peng, Xi",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.32",
doi = "10.18653/v1/2022.findings-emnlp.32",
pages = "460--470",
abstract = "Although explainable artificial intelligence (XAI) has achieved remarkable developments in recent years, there are few efforts have been devoted to the following problems, namely, i) how to develop an explainable method that could explain the black-box in a model-agnostic way? and ii) how to improve the performance and interpretability of the black-box using such explanations instead of pre-collected important attributions? To explore the potential solution, we propose a model-agnostic explanation method termed as Sparse Contrastive Coding (SCC) and verify its effectiveness in text classification and natural language inference. In brief, SCC explains the feature attributions which characterize the importance of words based on the hidden states of each layer of the model. With such word-level explainability, SCC adaptively divides the input sentences into foregrounds and backgrounds in terms of task relevance. Through maximizing the similarity between the foregrounds and input sentences while minimizing the similarity between the backgrounds and input sentences, SSC employs a supervised contrastive learning loss to boost the interpretability and performance of the model. Extensive experiments show the superiority of our method over five state-of-the-art methods in terms of interpretability and classification measurements. The code is available at \url{https://pengxi.me}.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="liu-etal-2022-improve">
<titleInfo>
<title>Improve Interpretability of Neural Networks via Sparse Contrastive Coding</title>
</titleInfo>
<name type="personal">
<namePart type="given">Junhong</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yijie</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Liang</namePart>
<namePart type="family">Jiang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jia</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zujie</namePart>
<namePart type="family">Wen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xi</namePart>
<namePart type="family">Peng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2022</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Although explainable artificial intelligence (XAI) has achieved remarkable developments in recent years, there are few efforts have been devoted to the following problems, namely, i) how to develop an explainable method that could explain the black-box in a model-agnostic way? and ii) how to improve the performance and interpretability of the black-box using such explanations instead of pre-collected important attributions? To explore the potential solution, we propose a model-agnostic explanation method termed as Sparse Contrastive Coding (SCC) and verify its effectiveness in text classification and natural language inference. In brief, SCC explains the feature attributions which characterize the importance of words based on the hidden states of each layer of the model. With such word-level explainability, SCC adaptively divides the input sentences into foregrounds and backgrounds in terms of task relevance. Through maximizing the similarity between the foregrounds and input sentences while minimizing the similarity between the backgrounds and input sentences, SSC employs a supervised contrastive learning loss to boost the interpretability and performance of the model. Extensive experiments show the superiority of our method over five state-of-the-art methods in terms of interpretability and classification measurements. The code is available at https://pengxi.me.</abstract>
<identifier type="citekey">liu-etal-2022-improve</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.32</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.32</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>460</start>
<end>470</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Improve Interpretability of Neural Networks via Sparse Contrastive Coding
%A Liu, Junhong
%A Lin, Yijie
%A Jiang, Liang
%A Liu, Jia
%A Wen, Zujie
%A Peng, Xi
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F liu-etal-2022-improve
%X Although explainable artificial intelligence (XAI) has achieved remarkable developments in recent years, there are few efforts have been devoted to the following problems, namely, i) how to develop an explainable method that could explain the black-box in a model-agnostic way? and ii) how to improve the performance and interpretability of the black-box using such explanations instead of pre-collected important attributions? To explore the potential solution, we propose a model-agnostic explanation method termed as Sparse Contrastive Coding (SCC) and verify its effectiveness in text classification and natural language inference. In brief, SCC explains the feature attributions which characterize the importance of words based on the hidden states of each layer of the model. With such word-level explainability, SCC adaptively divides the input sentences into foregrounds and backgrounds in terms of task relevance. Through maximizing the similarity between the foregrounds and input sentences while minimizing the similarity between the backgrounds and input sentences, SSC employs a supervised contrastive learning loss to boost the interpretability and performance of the model. Extensive experiments show the superiority of our method over five state-of-the-art methods in terms of interpretability and classification measurements. The code is available at https://pengxi.me.
%R 10.18653/v1/2022.findings-emnlp.32
%U https://aclanthology.org/2022.findings-emnlp.32
%U https://doi.org/10.18653/v1/2022.findings-emnlp.32
%P 460-470
Markdown (Informal)
[Improve Interpretability of Neural Networks via Sparse Contrastive Coding](https://aclanthology.org/2022.findings-emnlp.32) (Liu et al., Findings 2022)
ACL