@inproceedings{deng-etal-2022-valcat,
title = "{V}al{CAT}: Variable-Length Contextualized Adversarial Transformations Using Encoder-Decoder Language Model",
author = "Deng, Chuyun and
Liu, Mingxuan and
Qin, Yue and
Zhang, Jia and
Duan, Hai-Xin and
Sun, Donghong",
editor = "Carpuat, Marine and
de Marneffe, Marie-Catherine and
Meza Ruiz, Ivan Vladimir",
booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies",
month = jul,
year = "2022",
address = "Seattle, United States",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.naacl-main.125",
doi = "10.18653/v1/2022.naacl-main.125",
pages = "1735--1746",
abstract = "Adversarial texts help explore vulnerabilities in language models, improve model robustness, and explain their working mechanisms. However, existing word-level attack methods trap in a one-to-one attack pattern, i.e., only a single word can be modified in one transformation round, and they ignore the interactions between several consecutive words. In this paper, we propose ValCAT, a black-box attack framework that misleads the language model by applying variable-length contextualized transformations to the original text. Compared to word-level methods, ValCAT expands the basic units of perturbation from single words to spans composed of multiple consecutive words, enhancing the perturbation capability. Experiments show that our method outperforms state-of-the-art methods in terms of attack success rate, perplexity, and semantic similarity on several classification tasks and inference tasks. The comprehensive human evaluation demonstrates that ValCAT has a significant advantage in ensuring the fluency of the adversarial examples and achieves better semantic consistency. We release the code at \url{https://github.com/linerxliner/ValCAT}.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="deng-etal-2022-valcat">
<titleInfo>
<title>ValCAT: Variable-Length Contextualized Adversarial Transformations Using Encoder-Decoder Language Model</title>
</titleInfo>
<name type="personal">
<namePart type="given">Chuyun</namePart>
<namePart type="family">Deng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingxuan</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Qin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jia</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hai-Xin</namePart>
<namePart type="family">Duan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Donghong</namePart>
<namePart type="family">Sun</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-07</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marine</namePart>
<namePart type="family">Carpuat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marie-Catherine</namePart>
<namePart type="family">de Marneffe</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Ivan</namePart>
<namePart type="given">Vladimir</namePart>
<namePart type="family">Meza Ruiz</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Seattle, United States</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Adversarial texts help explore vulnerabilities in language models, improve model robustness, and explain their working mechanisms. However, existing word-level attack methods trap in a one-to-one attack pattern, i.e., only a single word can be modified in one transformation round, and they ignore the interactions between several consecutive words. In this paper, we propose ValCAT, a black-box attack framework that misleads the language model by applying variable-length contextualized transformations to the original text. Compared to word-level methods, ValCAT expands the basic units of perturbation from single words to spans composed of multiple consecutive words, enhancing the perturbation capability. Experiments show that our method outperforms state-of-the-art methods in terms of attack success rate, perplexity, and semantic similarity on several classification tasks and inference tasks. The comprehensive human evaluation demonstrates that ValCAT has a significant advantage in ensuring the fluency of the adversarial examples and achieves better semantic consistency. We release the code at https://github.com/linerxliner/ValCAT.</abstract>
<identifier type="citekey">deng-etal-2022-valcat</identifier>
<identifier type="doi">10.18653/v1/2022.naacl-main.125</identifier>
<location>
<url>https://aclanthology.org/2022.naacl-main.125</url>
</location>
<part>
<date>2022-07</date>
<extent unit="page">
<start>1735</start>
<end>1746</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T ValCAT: Variable-Length Contextualized Adversarial Transformations Using Encoder-Decoder Language Model
%A Deng, Chuyun
%A Liu, Mingxuan
%A Qin, Yue
%A Zhang, Jia
%A Duan, Hai-Xin
%A Sun, Donghong
%Y Carpuat, Marine
%Y de Marneffe, Marie-Catherine
%Y Meza Ruiz, Ivan Vladimir
%S Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
%D 2022
%8 July
%I Association for Computational Linguistics
%C Seattle, United States
%F deng-etal-2022-valcat
%X Adversarial texts help explore vulnerabilities in language models, improve model robustness, and explain their working mechanisms. However, existing word-level attack methods trap in a one-to-one attack pattern, i.e., only a single word can be modified in one transformation round, and they ignore the interactions between several consecutive words. In this paper, we propose ValCAT, a black-box attack framework that misleads the language model by applying variable-length contextualized transformations to the original text. Compared to word-level methods, ValCAT expands the basic units of perturbation from single words to spans composed of multiple consecutive words, enhancing the perturbation capability. Experiments show that our method outperforms state-of-the-art methods in terms of attack success rate, perplexity, and semantic similarity on several classification tasks and inference tasks. The comprehensive human evaluation demonstrates that ValCAT has a significant advantage in ensuring the fluency of the adversarial examples and achieves better semantic consistency. We release the code at https://github.com/linerxliner/ValCAT.
%R 10.18653/v1/2022.naacl-main.125
%U https://aclanthology.org/2022.naacl-main.125
%U https://doi.org/10.18653/v1/2022.naacl-main.125
%P 1735-1746
Markdown (Informal)
[ValCAT: Variable-Length Contextualized Adversarial Transformations Using Encoder-Decoder Language Model](https://aclanthology.org/2022.naacl-main.125) (Deng et al., NAACL 2022)
ACL