@inproceedings{tran-etal-2023-impacts,
title = "The Impacts of Unanswerable Questions on the Robustness of Machine Reading Comprehension Models",
author = "Tran, Son Quoc and
Do, Phong Nguyen-Thuan and
Le, Uyen and
Kretchmar, Matt",
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.113",
doi = "10.18653/v1/2023.eacl-main.113",
pages = "1543--1557",
abstract = "Pretrained language models have achieved super-human performances on many Machine Reading Comprehension (MRC) benchmarks. Nevertheless, their relative inability to defend against adversarial attacks has spurred skepticism about their natural language understanding. In this paper, we ask whether training with unanswerable questions in SQuAD 2.0 can help improve the robustness of MRC models against adversarial attacks. To explore that question, we fine-tune three state-of-the-art language models on either SQuAD 1.1 or SQuAD 2.0 and then evaluate their robustness under adversarial attacks. Our experiments reveal that current models fine-tuned on SQuAD 2.0 do not initially appear to be any more robust than ones fine-tuned on SQuAD 1.1, yet they reveal a measure of hidden robustness that can be leveraged to realize actual performance gains. Furthermore, we find that robustness of models fine-tuned on SQuAD 2.0 extends on additional out-of-domain datasets. Finally, we introduce a new adversarial attack to reveal of SQuAD 2.0 that current MRC models are learning.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="tran-etal-2023-impacts">
<titleInfo>
<title>The Impacts of Unanswerable Questions on the Robustness of Machine Reading Comprehension Models</title>
</titleInfo>
<name type="personal">
<namePart type="given">Son</namePart>
<namePart type="given">Quoc</namePart>
<namePart type="family">Tran</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Phong</namePart>
<namePart type="given">Nguyen-Thuan</namePart>
<namePart type="family">Do</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Uyen</namePart>
<namePart type="family">Le</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Matt</namePart>
<namePart type="family">Kretchmar</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>Pretrained language models have achieved super-human performances on many Machine Reading Comprehension (MRC) benchmarks. Nevertheless, their relative inability to defend against adversarial attacks has spurred skepticism about their natural language understanding. In this paper, we ask whether training with unanswerable questions in SQuAD 2.0 can help improve the robustness of MRC models against adversarial attacks. To explore that question, we fine-tune three state-of-the-art language models on either SQuAD 1.1 or SQuAD 2.0 and then evaluate their robustness under adversarial attacks. Our experiments reveal that current models fine-tuned on SQuAD 2.0 do not initially appear to be any more robust than ones fine-tuned on SQuAD 1.1, yet they reveal a measure of hidden robustness that can be leveraged to realize actual performance gains. Furthermore, we find that robustness of models fine-tuned on SQuAD 2.0 extends on additional out-of-domain datasets. Finally, we introduce a new adversarial attack to reveal of SQuAD 2.0 that current MRC models are learning.</abstract>
<identifier type="citekey">tran-etal-2023-impacts</identifier>
<identifier type="doi">10.18653/v1/2023.eacl-main.113</identifier>
<location>
<url>https://aclanthology.org/2023.eacl-main.113</url>
</location>
<part>
<date>2023-05</date>
<extent unit="page">
<start>1543</start>
<end>1557</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T The Impacts of Unanswerable Questions on the Robustness of Machine Reading Comprehension Models
%A Tran, Son Quoc
%A Do, Phong Nguyen-Thuan
%A Le, Uyen
%A Kretchmar, Matt
%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 tran-etal-2023-impacts
%X Pretrained language models have achieved super-human performances on many Machine Reading Comprehension (MRC) benchmarks. Nevertheless, their relative inability to defend against adversarial attacks has spurred skepticism about their natural language understanding. In this paper, we ask whether training with unanswerable questions in SQuAD 2.0 can help improve the robustness of MRC models against adversarial attacks. To explore that question, we fine-tune three state-of-the-art language models on either SQuAD 1.1 or SQuAD 2.0 and then evaluate their robustness under adversarial attacks. Our experiments reveal that current models fine-tuned on SQuAD 2.0 do not initially appear to be any more robust than ones fine-tuned on SQuAD 1.1, yet they reveal a measure of hidden robustness that can be leveraged to realize actual performance gains. Furthermore, we find that robustness of models fine-tuned on SQuAD 2.0 extends on additional out-of-domain datasets. Finally, we introduce a new adversarial attack to reveal of SQuAD 2.0 that current MRC models are learning.
%R 10.18653/v1/2023.eacl-main.113
%U https://aclanthology.org/2023.eacl-main.113
%U https://doi.org/10.18653/v1/2023.eacl-main.113
%P 1543-1557
Markdown (Informal)
[The Impacts of Unanswerable Questions on the Robustness of Machine Reading Comprehension Models](https://aclanthology.org/2023.eacl-main.113) (Tran et al., EACL 2023)
ACL