@inproceedings{bayer-etal-2024-xai-attack,
title = "{XAI}-Attack: Utilizing Explainable {AI} to Find Incorrectly Learned Patterns for Black-Box Adversarial Example Creation",
author = {Bayer, Markus and
Neiczer, Markus and
Samsinger, Maximilian and
Buchhold, Bj{\"o}rn and
Reuter, Christian},
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.1542",
pages = "17725--17738",
abstract = "Adversarial examples, capable of misleading machine learning models into making erroneous predictions, pose significant risks in safety-critical domains such as crisis informatics, medicine, and autonomous driving. To counter this, we introduce a novel textual adversarial example method that identifies falsely learned word indicators by leveraging explainable AI methods as importance functions on incorrectly predicted instances, thus revealing and understanding the weaknesses of a model. To evaluate the effectiveness of our approach, we conduct a human and a transfer evaluation and propose a novel adversarial training evaluation setting for better robustness assessment. While outperforming current adversarial example and training methods, the results also show our method{'}s potential in facilitating the development of more resilient transformer models by detecting and rectifying biases and patterns in training data, showing baseline improvements of up to 23 percentage points in accuracy on adversarial tasks. The code of our approach is freely available for further exploration and use.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="bayer-etal-2024-xai-attack">
<titleInfo>
<title>XAI-Attack: Utilizing Explainable AI to Find Incorrectly Learned Patterns for Black-Box Adversarial Example Creation</title>
</titleInfo>
<name type="personal">
<namePart type="given">Markus</namePart>
<namePart type="family">Bayer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Markus</namePart>
<namePart type="family">Neiczer</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Maximilian</namePart>
<namePart type="family">Samsinger</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Björn</namePart>
<namePart type="family">Buchhold</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christian</namePart>
<namePart type="family">Reuter</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>Adversarial examples, capable of misleading machine learning models into making erroneous predictions, pose significant risks in safety-critical domains such as crisis informatics, medicine, and autonomous driving. To counter this, we introduce a novel textual adversarial example method that identifies falsely learned word indicators by leveraging explainable AI methods as importance functions on incorrectly predicted instances, thus revealing and understanding the weaknesses of a model. To evaluate the effectiveness of our approach, we conduct a human and a transfer evaluation and propose a novel adversarial training evaluation setting for better robustness assessment. While outperforming current adversarial example and training methods, the results also show our method’s potential in facilitating the development of more resilient transformer models by detecting and rectifying biases and patterns in training data, showing baseline improvements of up to 23 percentage points in accuracy on adversarial tasks. The code of our approach is freely available for further exploration and use.</abstract>
<identifier type="citekey">bayer-etal-2024-xai-attack</identifier>
<location>
<url>https://aclanthology.org/2024.lrec-main.1542</url>
</location>
<part>
<date>2024-05</date>
<extent unit="page">
<start>17725</start>
<end>17738</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T XAI-Attack: Utilizing Explainable AI to Find Incorrectly Learned Patterns for Black-Box Adversarial Example Creation
%A Bayer, Markus
%A Neiczer, Markus
%A Samsinger, Maximilian
%A Buchhold, Björn
%A Reuter, Christian
%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 bayer-etal-2024-xai-attack
%X Adversarial examples, capable of misleading machine learning models into making erroneous predictions, pose significant risks in safety-critical domains such as crisis informatics, medicine, and autonomous driving. To counter this, we introduce a novel textual adversarial example method that identifies falsely learned word indicators by leveraging explainable AI methods as importance functions on incorrectly predicted instances, thus revealing and understanding the weaknesses of a model. To evaluate the effectiveness of our approach, we conduct a human and a transfer evaluation and propose a novel adversarial training evaluation setting for better robustness assessment. While outperforming current adversarial example and training methods, the results also show our method’s potential in facilitating the development of more resilient transformer models by detecting and rectifying biases and patterns in training data, showing baseline improvements of up to 23 percentage points in accuracy on adversarial tasks. The code of our approach is freely available for further exploration and use.
%U https://aclanthology.org/2024.lrec-main.1542
%P 17725-17738
Markdown (Informal)
[XAI-Attack: Utilizing Explainable AI to Find Incorrectly Learned Patterns for Black-Box Adversarial Example Creation](https://aclanthology.org/2024.lrec-main.1542) (Bayer et al., LREC-COLING 2024)
ACL