@inproceedings{chen-etal-2024-taichi-improving,
title = "{T}ai{C}hi: Improving the Robustness of {NLP} Models by Seeking Common Ground While Reserving Differences",
author = "Chen, Huimin and
Wang, Chengyu and
Wang, Yanhao and
Chen, Cen and
Wang, Yinggui",
editor = "Calzolari, Nicoletta and
Kan, Min-Yen and
Hoste, Veronique and
Lenci, Alessandro and
Sakti, Sakriani and
Xue, Nianwen",
booktitle = "Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)",
month = may,
year = "2024",
address = "Torino, Italia",
publisher = "ELRA and ICCL",
url = "https://aclanthology.org/2024.lrec-main.1351",
pages = "15542--15551",
abstract = "Recent studies have shown that Pre-trained Language Models (PLMs) are vulnerable to adversarial examples, crafted by introducing human-imperceptible perturbations to clean examples to deceive the models. This vulnerability stems from the divergence in the data distributions of clean and adversarial examples. Therefore, addressing this issue involves teaching the model to diminish the differences between the two types of samples and to focus more on their similarities. To this end, we propose a novel approach named \textit{TaiChi} that employs a Siamese network architecture. Specifically, it consists of two sub-networks sharing the same structure but trained on clean and adversarial samples, respectively, and uses a contrastive learning strategy to encourage the generation of similar language representations for both kinds of samples. Furthermore, it utilizes the Kullback-Leibler (KL) divergence loss to enhance the consistency in the predictive behavior of the two sub-networks. Extensive experiments across three widely used datasets demonstrate that \textit{TaiChi} achieves superior trade-offs between robustness to adversarial attacks at token and character levels and accuracy on clean examples compared to previous defense methods. Our code and data are publicly available at \url{https://github.com/sai4july/TaiChi}.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="chen-etal-2024-taichi-improving">
<titleInfo>
<title>TaiChi: Improving the Robustness of NLP Models by Seeking Common Ground While Reserving Differences</title>
</titleInfo>
<name type="personal">
<namePart type="given">Huimin</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Chengyu</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanhao</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Cen</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yinggui</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)</title>
</titleInfo>
<name type="personal">
<namePart type="given">Nicoletta</namePart>
<namePart type="family">Calzolari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Min-Yen</namePart>
<namePart type="family">Kan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Veronique</namePart>
<namePart type="family">Hoste</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Alessandro</namePart>
<namePart type="family">Lenci</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sakriani</namePart>
<namePart type="family">Sakti</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nianwen</namePart>
<namePart type="family">Xue</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>ELRA and ICCL</publisher>
<place>
<placeTerm type="text">Torino, Italia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Recent studies have shown that Pre-trained Language Models (PLMs) are vulnerable to adversarial examples, crafted by introducing human-imperceptible perturbations to clean examples to deceive the models. This vulnerability stems from the divergence in the data distributions of clean and adversarial examples. Therefore, addressing this issue involves teaching the model to diminish the differences between the two types of samples and to focus more on their similarities. To this end, we propose a novel approach named TaiChi that employs a Siamese network architecture. Specifically, it consists of two sub-networks sharing the same structure but trained on clean and adversarial samples, respectively, and uses a contrastive learning strategy to encourage the generation of similar language representations for both kinds of samples. Furthermore, it utilizes the Kullback-Leibler (KL) divergence loss to enhance the consistency in the predictive behavior of the two sub-networks. Extensive experiments across three widely used datasets demonstrate that TaiChi achieves superior trade-offs between robustness to adversarial attacks at token and character levels and accuracy on clean examples compared to previous defense methods. Our code and data are publicly available at https://github.com/sai4july/TaiChi.</abstract>
<identifier type="citekey">chen-etal-2024-taichi-improving</identifier>
<location>
<url>https://aclanthology.org/2024.lrec-main.1351</url>
</location>
<part>
<date>2024-05</date>
<extent unit="page">
<start>15542</start>
<end>15551</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T TaiChi: Improving the Robustness of NLP Models by Seeking Common Ground While Reserving Differences
%A Chen, Huimin
%A Wang, Chengyu
%A Wang, Yanhao
%A Chen, Cen
%A Wang, Yinggui
%Y Calzolari, Nicoletta
%Y Kan, Min-Yen
%Y Hoste, Veronique
%Y Lenci, Alessandro
%Y Sakti, Sakriani
%Y Xue, Nianwen
%S Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024)
%D 2024
%8 May
%I ELRA and ICCL
%C Torino, Italia
%F chen-etal-2024-taichi-improving
%X Recent studies have shown that Pre-trained Language Models (PLMs) are vulnerable to adversarial examples, crafted by introducing human-imperceptible perturbations to clean examples to deceive the models. This vulnerability stems from the divergence in the data distributions of clean and adversarial examples. Therefore, addressing this issue involves teaching the model to diminish the differences between the two types of samples and to focus more on their similarities. To this end, we propose a novel approach named TaiChi that employs a Siamese network architecture. Specifically, it consists of two sub-networks sharing the same structure but trained on clean and adversarial samples, respectively, and uses a contrastive learning strategy to encourage the generation of similar language representations for both kinds of samples. Furthermore, it utilizes the Kullback-Leibler (KL) divergence loss to enhance the consistency in the predictive behavior of the two sub-networks. Extensive experiments across three widely used datasets demonstrate that TaiChi achieves superior trade-offs between robustness to adversarial attacks at token and character levels and accuracy on clean examples compared to previous defense methods. Our code and data are publicly available at https://github.com/sai4july/TaiChi.
%U https://aclanthology.org/2024.lrec-main.1351
%P 15542-15551
Markdown (Informal)
[TaiChi: Improving the Robustness of NLP Models by Seeking Common Ground While Reserving Differences](https://aclanthology.org/2024.lrec-main.1351) (Chen et al., LREC-COLING 2024)
ACL