@inproceedings{pamies-etal-2023-weakly,
title = "A weakly supervised textual entailment approach to zero-shot text classification",
author = "P{\`a}mies, Marc and
Llop, Joan and
Multari, Francesco and
Duran-Silva, Nicolau and
Parra-Rojas, C{\'e}sar and
Gonzalez-Agirre, Aitor and
Massucci, Francesco Alessandro and
Villegas, Marta",
editor = "Vlachos, Andreas and
Augenstein, Isabelle",
booktitle = "Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics",
month = may,
year = "2023",
address = "Dubrovnik, Croatia",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2023.eacl-main.22",
doi = "10.18653/v1/2023.eacl-main.22",
pages = "286--296",
abstract = "Zero-shot text classification is a widely studied task that deals with a lack of annotated data. The most common approach is to reformulate it as a textual entailment problem, enabling classification into unseen classes. This work explores an effective approach that trains on a weakly supervised dataset generated from traditional classification data. We empirically study the relation between the performance of the entailment task, which is used as a proxy, and the target zero-shot text classification task. Our findings reveal that there is no linear correlation between both tasks, to the extent that it can be detrimental to lengthen the fine-tuning process even when the model is still learning, and propose a straightforward method to stop training on time. As a proof of concept, we introduce a domain-specific zero-shot text classifier that was trained on Microsoft Academic Graph data. The model, called SCIroShot, achieves state-of-the-art performance in the scientific domain and competitive results in other areas. Both the model and evaluation benchmark are publicly available on HuggingFace and GitHub.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="pamies-etal-2023-weakly">
<titleInfo>
<title>A weakly supervised textual entailment approach to zero-shot text classification</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marc</namePart>
<namePart type="family">Pàmies</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joan</namePart>
<namePart type="family">Llop</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Francesco</namePart>
<namePart type="family">Multari</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Nicolau</namePart>
<namePart type="family">Duran-Silva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">César</namePart>
<namePart type="family">Parra-Rojas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Aitor</namePart>
<namePart type="family">Gonzalez-Agirre</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Francesco</namePart>
<namePart type="given">Alessandro</namePart>
<namePart type="family">Massucci</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Marta</namePart>
<namePart type="family">Villegas</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2023-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics</title>
</titleInfo>
<name type="personal">
<namePart type="given">Andreas</namePart>
<namePart type="family">Vlachos</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Isabelle</namePart>
<namePart type="family">Augenstein</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Dubrovnik, Croatia</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>Zero-shot text classification is a widely studied task that deals with a lack of annotated data. The most common approach is to reformulate it as a textual entailment problem, enabling classification into unseen classes. This work explores an effective approach that trains on a weakly supervised dataset generated from traditional classification data. We empirically study the relation between the performance of the entailment task, which is used as a proxy, and the target zero-shot text classification task. Our findings reveal that there is no linear correlation between both tasks, to the extent that it can be detrimental to lengthen the fine-tuning process even when the model is still learning, and propose a straightforward method to stop training on time. As a proof of concept, we introduce a domain-specific zero-shot text classifier that was trained on Microsoft Academic Graph data. The model, called SCIroShot, achieves state-of-the-art performance in the scientific domain and competitive results in other areas. Both the model and evaluation benchmark are publicly available on HuggingFace and GitHub.</abstract>
<identifier type="citekey">pamies-etal-2023-weakly</identifier>
<identifier type="doi">10.18653/v1/2023.eacl-main.22</identifier>
<location>
<url>https://aclanthology.org/2023.eacl-main.22</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>286</start>
<end>296</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T A weakly supervised textual entailment approach to zero-shot text classification
%A Pàmies, Marc
%A Llop, Joan
%A Multari, Francesco
%A Duran-Silva, Nicolau
%A Parra-Rojas, César
%A Gonzalez-Agirre, Aitor
%A Massucci, Francesco Alessandro
%A Villegas, Marta
%Y Vlachos, Andreas
%Y Augenstein, Isabelle
%S Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics
%D 2023
%8 May
%I Association for Computational Linguistics
%C Dubrovnik, Croatia
%F pamies-etal-2023-weakly
%X Zero-shot text classification is a widely studied task that deals with a lack of annotated data. The most common approach is to reformulate it as a textual entailment problem, enabling classification into unseen classes. This work explores an effective approach that trains on a weakly supervised dataset generated from traditional classification data. We empirically study the relation between the performance of the entailment task, which is used as a proxy, and the target zero-shot text classification task. Our findings reveal that there is no linear correlation between both tasks, to the extent that it can be detrimental to lengthen the fine-tuning process even when the model is still learning, and propose a straightforward method to stop training on time. As a proof of concept, we introduce a domain-specific zero-shot text classifier that was trained on Microsoft Academic Graph data. The model, called SCIroShot, achieves state-of-the-art performance in the scientific domain and competitive results in other areas. Both the model and evaluation benchmark are publicly available on HuggingFace and GitHub.
%R 10.18653/v1/2023.eacl-main.22
%U https://aclanthology.org/2023.eacl-main.22
%U https://doi.org/10.18653/v1/2023.eacl-main.22
%P 286-296
Markdown (Informal)
[A weakly supervised textual entailment approach to zero-shot text classification](https://aclanthology.org/2023.eacl-main.22) (Pàmies et al., EACL 2023)
ACL
- Marc Pàmies, Joan Llop, Francesco Multari, Nicolau Duran-Silva, César Parra-Rojas, Aitor Gonzalez-Agirre, Francesco Alessandro Massucci, and Marta Villegas. 2023. A weakly supervised textual entailment approach to zero-shot text classification. In Proceedings of the 17th Conference of the European Chapter of the Association for Computational Linguistics, pages 286–296, Dubrovnik, Croatia. Association for Computational Linguistics.