@inproceedings{si-etal-2022-language,
title = "Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in {VQA}",
author = "Si, Qingyi and
Meng, Fandong and
Zheng, Mingyu and
Lin, Zheng and
Liu, Yuanxin and
Fu, Peng and
Cao, Yanan and
Wang, Weiping and
Zhou, Jie",
editor = "Goldberg, Yoav and
Kozareva, Zornitsa and
Zhang, Yue",
booktitle = "Findings of the Association for Computational Linguistics: EMNLP 2022",
month = dec,
year = "2022",
address = "Abu Dhabi, United Arab Emirates",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2022.findings-emnlp.271",
doi = "10.18653/v1/2022.findings-emnlp.271",
pages = "3698--3712",
abstract = "Visual Question Answering (VQA) models are prone to learn the shortcut solution formed by dataset biases rather than the intended solution. To evaluate the VQA models{'} reasoning ability beyond shortcut learning, the VQA-CP v2 dataset introduces a distribution shift between the training and test set given a question type. In this way, the model cannot use the training set shortcut (from question type to answer) to perform well on the test set. However, VQA-CP v2 only considers one type of shortcut and thus still cannot guarantee that the model relies on the intended solution rather than a solution specific to this shortcut. To overcome this limitation, we propose a new dataset that considers varying types of shortcuts by constructing different distribution shifts in multiple OOD test sets. In addition, we overcome the three troubling practices in the use of VQA-CP v2, e.g., selecting models using OOD test sets, and further standardize OOD evaluation procedure. Our benchmark provides a more rigorous and comprehensive testbed for shortcut learning in VQA. We benchmark recent methods and find that methods specifically designed for particular shortcuts fail to simultaneously generalize to our varying OOD test sets. We also systematically study the varying shortcuts and provide several valuable findings, which may promote the exploration of shortcut learning in VQA.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="si-etal-2022-language">
<titleInfo>
<title>Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in VQA</title>
</titleInfo>
<name type="personal">
<namePart type="given">Qingyi</namePart>
<namePart type="family">Si</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Fandong</namePart>
<namePart type="family">Meng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Mingyu</namePart>
<namePart type="family">Zheng</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Zheng</namePart>
<namePart type="family">Lin</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yuanxin</namePart>
<namePart type="family">Liu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Peng</namePart>
<namePart type="family">Fu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Yanan</namePart>
<namePart type="family">Cao</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Weiping</namePart>
<namePart type="family">Wang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Jie</namePart>
<namePart type="family">Zhou</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>Findings of the Association for Computational Linguistics: EMNLP 2022</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>Visual Question Answering (VQA) models are prone to learn the shortcut solution formed by dataset biases rather than the intended solution. To evaluate the VQA models’ reasoning ability beyond shortcut learning, the VQA-CP v2 dataset introduces a distribution shift between the training and test set given a question type. In this way, the model cannot use the training set shortcut (from question type to answer) to perform well on the test set. However, VQA-CP v2 only considers one type of shortcut and thus still cannot guarantee that the model relies on the intended solution rather than a solution specific to this shortcut. To overcome this limitation, we propose a new dataset that considers varying types of shortcuts by constructing different distribution shifts in multiple OOD test sets. In addition, we overcome the three troubling practices in the use of VQA-CP v2, e.g., selecting models using OOD test sets, and further standardize OOD evaluation procedure. Our benchmark provides a more rigorous and comprehensive testbed for shortcut learning in VQA. We benchmark recent methods and find that methods specifically designed for particular shortcuts fail to simultaneously generalize to our varying OOD test sets. We also systematically study the varying shortcuts and provide several valuable findings, which may promote the exploration of shortcut learning in VQA.</abstract>
<identifier type="citekey">si-etal-2022-language</identifier>
<identifier type="doi">10.18653/v1/2022.findings-emnlp.271</identifier>
<location>
<url>https://aclanthology.org/2022.findings-emnlp.271</url>
</location>
<part>
<date>2022-12</date>
<extent unit="page">
<start>3698</start>
<end>3712</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in VQA
%A Si, Qingyi
%A Meng, Fandong
%A Zheng, Mingyu
%A Lin, Zheng
%A Liu, Yuanxin
%A Fu, Peng
%A Cao, Yanan
%A Wang, Weiping
%A Zhou, Jie
%Y Goldberg, Yoav
%Y Kozareva, Zornitsa
%Y Zhang, Yue
%S Findings of the Association for Computational Linguistics: EMNLP 2022
%D 2022
%8 December
%I Association for Computational Linguistics
%C Abu Dhabi, United Arab Emirates
%F si-etal-2022-language
%X Visual Question Answering (VQA) models are prone to learn the shortcut solution formed by dataset biases rather than the intended solution. To evaluate the VQA models’ reasoning ability beyond shortcut learning, the VQA-CP v2 dataset introduces a distribution shift between the training and test set given a question type. In this way, the model cannot use the training set shortcut (from question type to answer) to perform well on the test set. However, VQA-CP v2 only considers one type of shortcut and thus still cannot guarantee that the model relies on the intended solution rather than a solution specific to this shortcut. To overcome this limitation, we propose a new dataset that considers varying types of shortcuts by constructing different distribution shifts in multiple OOD test sets. In addition, we overcome the three troubling practices in the use of VQA-CP v2, e.g., selecting models using OOD test sets, and further standardize OOD evaluation procedure. Our benchmark provides a more rigorous and comprehensive testbed for shortcut learning in VQA. We benchmark recent methods and find that methods specifically designed for particular shortcuts fail to simultaneously generalize to our varying OOD test sets. We also systematically study the varying shortcuts and provide several valuable findings, which may promote the exploration of shortcut learning in VQA.
%R 10.18653/v1/2022.findings-emnlp.271
%U https://aclanthology.org/2022.findings-emnlp.271
%U https://doi.org/10.18653/v1/2022.findings-emnlp.271
%P 3698-3712
Markdown (Informal)
[Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in VQA](https://aclanthology.org/2022.findings-emnlp.271) (Si et al., Findings 2022)
ACL
- Qingyi Si, Fandong Meng, Mingyu Zheng, Zheng Lin, Yuanxin Liu, Peng Fu, Yanan Cao, Weiping Wang, and Jie Zhou. 2022. Language Prior Is Not the Only Shortcut: A Benchmark for Shortcut Learning in VQA. In Findings of the Association for Computational Linguistics: EMNLP 2022, pages 3698–3712, Abu Dhabi, United Arab Emirates. Association for Computational Linguistics.