@inproceedings{paul-panenghat-etal-2020-towards,
title = "Towards the Necessity for Debiasing Natural Language Inference Datasets",
author = "Paul Panenghat, Mithun and
Suntwal, Sandeep and
Rafique, Faiz and
Sharp, Rebecca and
Surdeanu, Mihai",
editor = "Calzolari, Nicoletta and
B{\'e}chet, Fr{\'e}d{\'e}ric and
Blache, Philippe and
Choukri, Khalid and
Cieri, Christopher and
Declerck, Thierry and
Goggi, Sara and
Isahara, Hitoshi and
Maegaard, Bente and
Mariani, Joseph and
Mazo, H{\'e}l{\`e}ne and
Moreno, Asuncion and
Odijk, Jan and
Piperidis, Stelios",
booktitle = "Proceedings of the Twelfth Language Resources and Evaluation Conference",
month = may,
year = "2020",
address = "Marseille, France",
publisher = "European Language Resources Association",
url = "https://aclanthology.org/2020.lrec-1.850",
pages = "6883--6888",
abstract = "Modeling natural language inference is a challenging task. With large annotated data sets available it has now become feasible to train complex neural network based inference methods which achieve state of the art performance. However, it has been shown that these models also learn from the subtle biases inherent in these datasets (CITATION). In this work we explore two techniques for delexicalization that modify the datasets in such a way that we can control the importance that neural-network based methods place on lexical entities. We demonstrate that the proposed methods not only maintain the performance in-domain but also improve performance in some out-of-domain settings. For example, when using the delexicalized version of the FEVER dataset, the in-domain performance of a state of the art neural network method dropped only by 1.12{\%} while its out-of-domain performance on the FNC dataset improved by 4.63{\%}. We release the delexicalized versions of three common datasets used in natural language inference. These datasets are delexicalized using two methods: one which replaces the lexical entities in an overlap-aware manner, and a second, which additionally incorporates semantic lifting of nouns and verbs to their WordNet hypernym synsets",
language = "English",
ISBN = "979-10-95546-34-4",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="paul-panenghat-etal-2020-towards">
<titleInfo>
<title>Towards the Necessity for Debiasing Natural Language Inference Datasets</title>
</titleInfo>
<name type="personal">
<namePart type="given">Mithun</namePart>
<namePart type="family">Paul Panenghat</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sandeep</namePart>
<namePart type="family">Suntwal</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Faiz</namePart>
<namePart type="family">Rafique</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Rebecca</namePart>
<namePart type="family">Sharp</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mihai</namePart>
<namePart type="family">Surdeanu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2020-05</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<language>
<languageTerm type="text">English</languageTerm>
<languageTerm type="code" authority="iso639-2b">eng</languageTerm>
</language>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the Twelfth Language Resources and Evaluation Conference</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">Frédéric</namePart>
<namePart type="family">Béchet</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Philippe</namePart>
<namePart type="family">Blache</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Khalid</namePart>
<namePart type="family">Choukri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Christopher</namePart>
<namePart type="family">Cieri</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Thierry</namePart>
<namePart type="family">Declerck</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Sara</namePart>
<namePart type="family">Goggi</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hitoshi</namePart>
<namePart type="family">Isahara</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Bente</namePart>
<namePart type="family">Maegaard</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Joseph</namePart>
<namePart type="family">Mariani</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Hélène</namePart>
<namePart type="family">Mazo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Asuncion</namePart>
<namePart type="family">Moreno</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jan</namePart>
<namePart type="family">Odijk</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Stelios</namePart>
<namePart type="family">Piperidis</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>European Language Resources Association</publisher>
<place>
<placeTerm type="text">Marseille, France</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
<identifier type="isbn">979-10-95546-34-4</identifier>
</relatedItem>
<abstract>Modeling natural language inference is a challenging task. With large annotated data sets available it has now become feasible to train complex neural network based inference methods which achieve state of the art performance. However, it has been shown that these models also learn from the subtle biases inherent in these datasets (CITATION). In this work we explore two techniques for delexicalization that modify the datasets in such a way that we can control the importance that neural-network based methods place on lexical entities. We demonstrate that the proposed methods not only maintain the performance in-domain but also improve performance in some out-of-domain settings. For example, when using the delexicalized version of the FEVER dataset, the in-domain performance of a state of the art neural network method dropped only by 1.12% while its out-of-domain performance on the FNC dataset improved by 4.63%. We release the delexicalized versions of three common datasets used in natural language inference. These datasets are delexicalized using two methods: one which replaces the lexical entities in an overlap-aware manner, and a second, which additionally incorporates semantic lifting of nouns and verbs to their WordNet hypernym synsets</abstract>
<identifier type="citekey">paul-panenghat-etal-2020-towards</identifier>
<location>
<url>https://aclanthology.org/2020.lrec-1.850</url>
</location>
<part>
<date>2020-05</date>
<extent unit="page">
<start>6883</start>
<end>6888</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Towards the Necessity for Debiasing Natural Language Inference Datasets
%A Paul Panenghat, Mithun
%A Suntwal, Sandeep
%A Rafique, Faiz
%A Sharp, Rebecca
%A Surdeanu, Mihai
%Y Calzolari, Nicoletta
%Y Béchet, Frédéric
%Y Blache, Philippe
%Y Choukri, Khalid
%Y Cieri, Christopher
%Y Declerck, Thierry
%Y Goggi, Sara
%Y Isahara, Hitoshi
%Y Maegaard, Bente
%Y Mariani, Joseph
%Y Mazo, Hélène
%Y Moreno, Asuncion
%Y Odijk, Jan
%Y Piperidis, Stelios
%S Proceedings of the Twelfth Language Resources and Evaluation Conference
%D 2020
%8 May
%I European Language Resources Association
%C Marseille, France
%@ 979-10-95546-34-4
%G English
%F paul-panenghat-etal-2020-towards
%X Modeling natural language inference is a challenging task. With large annotated data sets available it has now become feasible to train complex neural network based inference methods which achieve state of the art performance. However, it has been shown that these models also learn from the subtle biases inherent in these datasets (CITATION). In this work we explore two techniques for delexicalization that modify the datasets in such a way that we can control the importance that neural-network based methods place on lexical entities. We demonstrate that the proposed methods not only maintain the performance in-domain but also improve performance in some out-of-domain settings. For example, when using the delexicalized version of the FEVER dataset, the in-domain performance of a state of the art neural network method dropped only by 1.12% while its out-of-domain performance on the FNC dataset improved by 4.63%. We release the delexicalized versions of three common datasets used in natural language inference. These datasets are delexicalized using two methods: one which replaces the lexical entities in an overlap-aware manner, and a second, which additionally incorporates semantic lifting of nouns and verbs to their WordNet hypernym synsets
%U https://aclanthology.org/2020.lrec-1.850
%P 6883-6888
Markdown (Informal)
[Towards the Necessity for Debiasing Natural Language Inference Datasets](https://aclanthology.org/2020.lrec-1.850) (Paul Panenghat et al., LREC 2020)
ACL