@inproceedings{akula-etal-2021-crossvqa,
title = "{C}ross{VQA}: Scalably Generating Benchmarks for Systematically Testing {VQA} Generalization",
author = "Akula, Arjun and
Changpinyo, Soravit and
Gong, Boqing and
Sharma, Piyush and
Zhu, Song-Chun and
Soricut, Radu",
editor = "Moens, Marie-Francine and
Huang, Xuanjing and
Specia, Lucia and
Yih, Scott Wen-tau",
booktitle = "Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing",
month = nov,
year = "2021",
address = "Online and Punta Cana, Dominican Republic",
publisher = "Association for Computational Linguistics",
url = "https://aclanthology.org/2021.emnlp-main.164",
doi = "10.18653/v1/2021.emnlp-main.164",
pages = "2148--2166",
abstract = "One challenge in evaluating visual question answering (VQA) models in the cross-dataset adaptation setting is that the distribution shifts are multi-modal, making it difficult to identify if it is the shifts in visual or language features that play a key role. In this paper, we propose a semi-automatic framework for generating disentangled shifts by introducing a controllable visual question-answer generation (VQAG) module that is capable of generating highly-relevant and diverse question-answer pairs with the desired dataset style. We use it to create CrossVQA, a collection of test splits for assessing VQA generalization based on the VQA2, VizWiz, and Open Images datasets. We provide an analysis of our generated datasets and demonstrate its utility by using them to evaluate several state-of-the-art VQA systems. One important finding is that the visual shifts in cross-dataset VQA matter more than the language shifts. More broadly, we present a scalable framework for systematically evaluating the machine with little human intervention.",
}
<?xml version="1.0" encoding="UTF-8"?>
<modsCollection xmlns="http://www.loc.gov/mods/v3">
<mods ID="akula-etal-2021-crossvqa">
<titleInfo>
<title>CrossVQA: Scalably Generating Benchmarks for Systematically Testing VQA Generalization</title>
</titleInfo>
<name type="personal">
<namePart type="given">Arjun</namePart>
<namePart type="family">Akula</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Soravit</namePart>
<namePart type="family">Changpinyo</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Boqing</namePart>
<namePart type="family">Gong</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Piyush</namePart>
<namePart type="family">Sharma</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Song-Chun</namePart>
<namePart type="family">Zhu</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Radu</namePart>
<namePart type="family">Soricut</namePart>
<role>
<roleTerm authority="marcrelator" type="text">author</roleTerm>
</role>
</name>
<originInfo>
<dateIssued>2021-11</dateIssued>
</originInfo>
<typeOfResource>text</typeOfResource>
<relatedItem type="host">
<titleInfo>
<title>Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing</title>
</titleInfo>
<name type="personal">
<namePart type="given">Marie-Francine</namePart>
<namePart type="family">Moens</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Xuanjing</namePart>
<namePart type="family">Huang</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Lucia</namePart>
<namePart type="family">Specia</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<name type="personal">
<namePart type="given">Scott</namePart>
<namePart type="given">Wen-tau</namePart>
<namePart type="family">Yih</namePart>
<role>
<roleTerm authority="marcrelator" type="text">editor</roleTerm>
</role>
</name>
<originInfo>
<publisher>Association for Computational Linguistics</publisher>
<place>
<placeTerm type="text">Online and Punta Cana, Dominican Republic</placeTerm>
</place>
</originInfo>
<genre authority="marcgt">conference publication</genre>
</relatedItem>
<abstract>One challenge in evaluating visual question answering (VQA) models in the cross-dataset adaptation setting is that the distribution shifts are multi-modal, making it difficult to identify if it is the shifts in visual or language features that play a key role. In this paper, we propose a semi-automatic framework for generating disentangled shifts by introducing a controllable visual question-answer generation (VQAG) module that is capable of generating highly-relevant and diverse question-answer pairs with the desired dataset style. We use it to create CrossVQA, a collection of test splits for assessing VQA generalization based on the VQA2, VizWiz, and Open Images datasets. We provide an analysis of our generated datasets and demonstrate its utility by using them to evaluate several state-of-the-art VQA systems. One important finding is that the visual shifts in cross-dataset VQA matter more than the language shifts. More broadly, we present a scalable framework for systematically evaluating the machine with little human intervention.</abstract>
<identifier type="citekey">akula-etal-2021-crossvqa</identifier>
<identifier type="doi">10.18653/v1/2021.emnlp-main.164</identifier>
<location>
<url>https://aclanthology.org/2021.emnlp-main.164</url>
</location>
<part>
<date>2021-11</date>
<extent unit="page">
<start>2148</start>
<end>2166</end>
</extent>
</part>
</mods>
</modsCollection>
%0 Conference Proceedings
%T CrossVQA: Scalably Generating Benchmarks for Systematically Testing VQA Generalization
%A Akula, Arjun
%A Changpinyo, Soravit
%A Gong, Boqing
%A Sharma, Piyush
%A Zhu, Song-Chun
%A Soricut, Radu
%Y Moens, Marie-Francine
%Y Huang, Xuanjing
%Y Specia, Lucia
%Y Yih, Scott Wen-tau
%S Proceedings of the 2021 Conference on Empirical Methods in Natural Language Processing
%D 2021
%8 November
%I Association for Computational Linguistics
%C Online and Punta Cana, Dominican Republic
%F akula-etal-2021-crossvqa
%X One challenge in evaluating visual question answering (VQA) models in the cross-dataset adaptation setting is that the distribution shifts are multi-modal, making it difficult to identify if it is the shifts in visual or language features that play a key role. In this paper, we propose a semi-automatic framework for generating disentangled shifts by introducing a controllable visual question-answer generation (VQAG) module that is capable of generating highly-relevant and diverse question-answer pairs with the desired dataset style. We use it to create CrossVQA, a collection of test splits for assessing VQA generalization based on the VQA2, VizWiz, and Open Images datasets. We provide an analysis of our generated datasets and demonstrate its utility by using them to evaluate several state-of-the-art VQA systems. One important finding is that the visual shifts in cross-dataset VQA matter more than the language shifts. More broadly, we present a scalable framework for systematically evaluating the machine with little human intervention.
%R 10.18653/v1/2021.emnlp-main.164
%U https://aclanthology.org/2021.emnlp-main.164
%U https://doi.org/10.18653/v1/2021.emnlp-main.164
%P 2148-2166
Markdown (Informal)
[CrossVQA: Scalably Generating Benchmarks for Systematically Testing VQA Generalization](https://aclanthology.org/2021.emnlp-main.164) (Akula et al., EMNLP 2021)
ACL