@inproceedings{rajaee-etal-2022-looking,
title = "Looking at the Overlooked: An Analysis on the Word-Overlap Bias in Natural Language Inference",
author = "Rajaee, Sara and
Yaghoobzadeh, Yadollah and
Pilehvar, Mohammad Taher",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.emnlp-main.725",
doi = "10.18653/v1/2022.emnlp-main.725",
pages = "10605--10616",
abstract = "It has been shown that NLI models are usually biased with respect to the word-overlap between the premise and the hypothesis, as they take this feature as a primary cue for predicting the entailment label. In this paper, we focus on an overlooked aspect of the overlap bias in the NLI models: the reverse word-overlap bias. Our experimental results demonstrate that current NLI systems are also highly biased towards the non-entailment label on instances with low overlap and that existing debiasing methods, which are reportedly successful on challenge datasets, are generally ineffective in addressing this category of bias. Through a set of analyses, we investigate the reasons for the emergence of the overlap bias and the role of minority examples in mitigating this bias. For the former, we find that the word overlap bias does not stem from pre-training, and in the latter, we observe that in contrast to the accepted assumption, eliminating minority examples does not affect the generalizability of debiasing methods with respect to the overlap bias.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="rajaee-etal-2022-looking">
<titleInfo>
<title>Looking at the Overlooked: An Analysis on the Word-Overlap Bias in Natural Language Inference</title>
</titleInfo>
<name type="personal">
<namePart type="given">Sara</namePart>
<namePart type="family">Rajaee</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yadollah</namePart>
<namePart type="family">Yaghoobzadeh</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mohammad</namePart>
<namePart type="given">Taher</namePart>
<namePart type="family">Pilehvar</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2022-12</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Yoav</namePart>
<namePart type="family">Goldberg</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zornitsa</namePart>
<namePart type="family">Kozareva</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yue</namePart>
<namePart type="family">Zhang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Abu Dhabi, United Arab Emirates</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>It has been shown that NLI models are usually biased with respect to the word-overlap between the premise and the hypothesis, as they take this feature as a primary cue for predicting the entailment label. In this paper, we focus on an overlooked aspect of the overlap bias in the NLI models: the reverse word-overlap bias. Our experimental results demonstrate that current NLI systems are also highly biased towards the non-entailment label on instances with low overlap and that existing debiasing methods, which are reportedly successful on challenge datasets, are generally ineffective in addressing this category of bias. Through a set of analyses, we investigate the reasons for the emergence of the overlap bias and the role of minority examples in mitigating this bias. For the former, we find that the word overlap bias does not stem from pre-training, and in the latter, we observe that in contrast to the accepted assumption, eliminating minority examples does not affect the generalizability of debiasing methods with respect to the overlap bias.</abstract>
<identifier type="citekey">rajaee-etal-2022-looking</identifier>
<identifier type="doi">10.18653/v1/2022.emnlp-main.725</identifier>
<location>
<url>https://aclanthology.org/2022.emnlp-main.725</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>10605</start>
<end>10616</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Looking at the Overlooked: An Analysis on the Word-Overlap Bias in Natural Language Inference
%A Rajaee, Sara
%A Yaghoobzadeh, Yadollah
%A Pilehvar, Mohammad Taher
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Proceedings of the 2022 Conference on Empirical Methods in Natural Language Processing
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F rajaee-etal-2022-looking
%X It has been shown that NLI models are usually biased with respect to the word-overlap between the premise and the hypothesis, as they take this feature as a primary cue for predicting the entailment label. In this paper, we focus on an overlooked aspect of the overlap bias in the NLI models: the reverse word-overlap bias. Our experimental results demonstrate that current NLI systems are also highly biased towards the non-entailment label on instances with low overlap and that existing debiasing methods, which are reportedly successful on challenge datasets, are generally ineffective in addressing this category of bias. Through a set of analyses, we investigate the reasons for the emergence of the overlap bias and the role of minority examples in mitigating this bias. For the former, we find that the word overlap bias does not stem from pre-training, and in the latter, we observe that in contrast to the accepted assumption, eliminating minority examples does not affect the generalizability of debiasing methods with respect to the overlap bias.
%R 10.18653/v1/2022.emnlp-main.725
%U https://aclanthology.org/2022.emnlp-main.725
%U https://doi.org/10.18653/v1/2022.emnlp-main.725
%P 10605-10616
Markdown (Informal)
[Looking at the Overlooked: An Analysis on the Word-Overlap Bias in Natural Language Inference](https://aclanthology.org/2022.emnlp-main.725) (Rajaee et al., EMNLP 2022)
ACL