@inproceedings{sourati-etal-2024-robust,
title = "Robust Text Classification: Analyzing Prototype-Based Networks",
author = "Sourati, Zhivar and
Deshpande, Darshan Girish and
Ilievski, Filip and
Gashteovski, Kiril and
Saralajew, Sascha",
editor = "Al-Onaizan, Yaser and
Bansal, Mohit and
Chen, Yun-Nung",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2024",
month = nov,
year = "2024",
address = "Miami, Florida, USA",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2024.findings-emnlp.745/",
doi = "10.18653/v1/2024.findings-emnlp.745",
pages = "12736--12757",
abstract = "Downstream applications often require text classification models to be accurate and robust. While the accuracy of state-of-the-art Language Models (LMs) approximates human performance, they often exhibit a drop in performance on real-world noisy data. This lack of robustness can be concerning, as even small perturbations in text, irrelevant to the target task, can cause classifiers to incorrectly change their predictions. A potential solution can be the family of Prototype-Based Networks (PBNs) that classifies examples based on their similarity to prototypical examples of a class (prototypes) and has been shown to be robust to noise for computer vision tasks. In this paper, we study whether the robustness properties of PBNs transfer to text classification tasks under both targeted and static adversarial attack settings. Our results show that PBNs, as a mere architectural variation of vanilla LMs, offer more robustness compared to vanilla LMs under both targeted and static settings. We showcase how PBNs' interpretability can help us understand PBNs' robustness properties. Finally, our ablation studies reveal the sensitivity of PBNs' robustness to the strictness of clustering and the number of prototypes in the training phase, as tighter clustering and a low number of prototypes result in less robust PBNs."
}<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="sourati-etal-2024-robust">
<titleInfo>
<title>Robust Text Classification: Analyzing Prototype-Based Networks</title>
</titleInfo>
<name type="personal">
<namePart type="given">Zhivar</namePart>
<namePart type="family">Sourati</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Darshan</namePart>
<namePart type="given">Girish</namePart>
<namePart type="family">Deshpande</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Filip</namePart>
<namePart type="family">Ilievski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Kiril</namePart>
<namePart type="family">Gashteovski</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sascha</namePart>
<namePart type="family">Saralajew</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2024-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Findings of the Association for Computational Linguistics: EMNLP 2024</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yaser</namePart>
<namePart type="family">Al-Onaizan</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohit</namePart>
<namePart type="family">Bansal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yun-Nung</namePart>
<namePart type="family">Chen</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Miami, Florida, USA</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Downstream applications often require text classification models to be accurate and robust. While the accuracy of state-of-the-art Language Models (LMs) approximates human performance, they often exhibit a drop in performance on real-world noisy data. This lack of robustness can be concerning, as even small perturbations in text, irrelevant to the target task, can cause classifiers to incorrectly change their predictions. A potential solution can be the family of Prototype-Based Networks (PBNs) that classifies examples based on their similarity to prototypical examples of a class (prototypes) and has been shown to be robust to noise for computer vision tasks. In this paper, we study whether the robustness properties of PBNs transfer to text classification tasks under both targeted and static adversarial attack settings. Our results show that PBNs, as a mere architectural variation of vanilla LMs, offer more robustness compared to vanilla LMs under both targeted and static settings. We showcase how PBNs’ interpretability can help us understand PBNs’ robustness properties. Finally, our ablation studies reveal the sensitivity of PBNs’ robustness to the strictness of clustering and the number of prototypes in the training phase, as tighter clustering and a low number of prototypes result in less robust PBNs.</abstract>
<identifier type="citekey">sourati-etal-2024-robust</identifier>
<identifier type="doi">10.18653/v1/2024.findings-emnlp.745</identifier>
<location>
<url>https://aclanthology.org/2024.findings-emnlp.745/</url>
</location>
<part>
<date>2024-11</date>
<extent unit="page">
<start>12736</start>
<end>12757</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Robust Text Classification: Analyzing Prototype-Based Networks
%A Sourati, Zhivar
%A Deshpande, Darshan Girish
%A Ilievski, Filip
%A Gashteovski, Kiril
%A Saralajew, Sascha
%Y Al-Onaizan, Yaser
%Y Bansal, Mohit
%Y Chen, Yun-Nung
%S Findings of the Association for Computational Linguistics: EMNLP 2024
%D 2024
%8 November
%I Association for Computational Linguistics
%C Miami, Florida, USA
%F sourati-etal-2024-robust
%X Downstream applications often require text classification models to be accurate and robust. While the accuracy of state-of-the-art Language Models (LMs) approximates human performance, they often exhibit a drop in performance on real-world noisy data. This lack of robustness can be concerning, as even small perturbations in text, irrelevant to the target task, can cause classifiers to incorrectly change their predictions. A potential solution can be the family of Prototype-Based Networks (PBNs) that classifies examples based on their similarity to prototypical examples of a class (prototypes) and has been shown to be robust to noise for computer vision tasks. In this paper, we study whether the robustness properties of PBNs transfer to text classification tasks under both targeted and static adversarial attack settings. Our results show that PBNs, as a mere architectural variation of vanilla LMs, offer more robustness compared to vanilla LMs under both targeted and static settings. We showcase how PBNs’ interpretability can help us understand PBNs’ robustness properties. Finally, our ablation studies reveal the sensitivity of PBNs’ robustness to the strictness of clustering and the number of prototypes in the training phase, as tighter clustering and a low number of prototypes result in less robust PBNs.
%R 10.18653/v1/2024.findings-emnlp.745
%U https://aclanthology.org/2024.findings-emnlp.745/
%U https://doi.org/10.18653/v1/2024.findings-emnlp.745
%P 12736-12757
Markdown (Informal)
[Robust Text Classification: Analyzing Prototype-Based Networks](https://aclanthology.org/2024.findings-emnlp.745/) (Sourati et al., Findings 2024)
ACL
- Zhivar Sourati, Darshan Girish Deshpande, Filip Ilievski, Kiril Gashteovski, and Sascha Saralajew. 2024. Robust Text Classification: Analyzing Prototype-Based Networks. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 12736–12757, Miami, Florida, USA. Association for Computational Linguistics.