Champion Solution for the WSDM2023 Toloka VQA Challenge

Champion Solution for the WSDM2023 Toloka VQA Challenge

In this report, we present our champion solution to the WSDM2023 Toloka Visual Question Answering (VQA) Challenge. Different from the common VQA and visual grounding (VG) tasks, this challenge involves a more complex scenario, i.e. inferring and locating the object implicitly specified by the given...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2023-02
Hauptverfasser: Gao, Shengyi, Chen, Zhe, Chen, Guo, Wang, Wenhai, Lu, Tong
Format: Artikel
Sprache:eng
Schlagworte:
Adapters
Questions
Task complexity
Visual tasks
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
container_end_page
container_issue
container_start_page
container_title arXiv.org
container_volume
creator Gao, Shengyi
Chen, Zhe
Chen, Guo
Wang, Wenhai
Lu, Tong
description In this report, we present our champion solution to the WSDM2023 Toloka Visual Question Answering (VQA) Challenge. Different from the common VQA and visual grounding (VG) tasks, this challenge involves a more complex scenario, i.e. inferring and locating the object implicitly specified by the given interrogative question. For this task, we leverage ViT-Adapter, a pre-training-free adapter network, to adapt multi-modal pre-trained Uni-Perceiver for better cross-modal localization. Our method ranks first on the leaderboard, achieving 77.5 and 76.347 IoU on public and private test sets, respectively. It shows that ViT-Adapter is also an effective paradigm for adapting the unified perception model to vision-language downstream tasks. Code and models will be released at https://github.com/czczup/ViT-Adapter/tree/main/wsdm2023.
format Article
fullrecord <record><control><sourceid>proquest</sourceid><recordid>TN_cdi_proquest_journals_2768910503</recordid><sourceformat>XML</sourceformat><sourcesystem>PC</sourcesystem><sourcerecordid>2768910503</sourcerecordid><originalsourceid>FETCH-proquest_journals_27689105033</originalsourceid><addsrcrecordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwd85IzC3IzM9TCM7PKS0BMdLyixRKMlIVwoNdfI0MjIwVQvJz8rMTFcICHRWAqnNyUvPSU3kYWNMSc4pTeaE0N4Oym2uIs4duQVF-YWlqcUl8Vn5pUR5QKt7I3MzC0tDA1MDYmDhVADjEM9k</addsrcrecordid><sourcetype>Aggregation Database</sourcetype><iscdi>true</iscdi><recordtype>article</recordtype><pqid>2768910503</pqid></control><display><type>article</type><title>Champion Solution for the WSDM2023 Toloka VQA Challenge</title><source>Free E- Journals</source><creator>Gao, Shengyi ; Chen, Zhe ; Chen, Guo ; Wang, Wenhai ; Lu, Tong</creator><creatorcontrib>Gao, Shengyi ; Chen, Zhe ; Chen, Guo ; Wang, Wenhai ; Lu, Tong</creatorcontrib><description>In this report, we present our champion solution to the WSDM2023 Toloka Visual Question Answering (VQA) Challenge. Different from the common VQA and visual grounding (VG) tasks, this challenge involves a more complex scenario, i.e. inferring and locating the object implicitly specified by the given interrogative question. For this task, we leverage ViT-Adapter, a pre-training-free adapter network, to adapt multi-modal pre-trained Uni-Perceiver for better cross-modal localization. Our method ranks first on the leaderboard, achieving 77.5 and 76.347 IoU on public and private test sets, respectively. It shows that ViT-Adapter is also an effective paradigm for adapting the unified perception model to vision-language downstream tasks. Code and models will be released at https://github.com/czczup/ViT-Adapter/tree/main/wsdm2023.</description><identifier>EISSN: 2331-8422</identifier><language>eng</language><publisher>Ithaca: Cornell University Library, arXiv.org</publisher><subject>Adapters ; Questions ; Task complexity ; Visual tasks</subject><ispartof>arXiv.org, 2023-02</ispartof><rights>2023. This work is published under http://arxiv.org/licenses/nonexclusive-distrib/1.0/ (the “License”). Notwithstanding the ProQuest Terms and Conditions, you may use this content in accordance with the terms of the License.</rights><oa>free_for_read</oa><woscitedreferencessubscribed>false</woscitedreferencessubscribed></display><links><openurl>$$Topenurl_article</openurl><openurlfulltext>$$Topenurlfull_article</openurlfulltext><thumbnail>$$Tsyndetics_thumb_exl</thumbnail><link.rule.ids>780,784</link.rule.ids></links><search><creatorcontrib>Gao, Shengyi</creatorcontrib><creatorcontrib>Chen, Zhe</creatorcontrib><creatorcontrib>Chen, Guo</creatorcontrib><creatorcontrib>Wang, Wenhai</creatorcontrib><creatorcontrib>Lu, Tong</creatorcontrib><title>Champion Solution for the WSDM2023 Toloka VQA Challenge</title><title>arXiv.org</title><description>In this report, we present our champion solution to the WSDM2023 Toloka Visual Question Answering (VQA) Challenge. Different from the common VQA and visual grounding (VG) tasks, this challenge involves a more complex scenario, i.e. inferring and locating the object implicitly specified by the given interrogative question. For this task, we leverage ViT-Adapter, a pre-training-free adapter network, to adapt multi-modal pre-trained Uni-Perceiver for better cross-modal localization. Our method ranks first on the leaderboard, achieving 77.5 and 76.347 IoU on public and private test sets, respectively. It shows that ViT-Adapter is also an effective paradigm for adapting the unified perception model to vision-language downstream tasks. Code and models will be released at https://github.com/czczup/ViT-Adapter/tree/main/wsdm2023.</description><subject>Adapters</subject><subject>Questions</subject><subject>Task complexity</subject><subject>Visual tasks</subject><issn>2331-8422</issn><fulltext>true</fulltext><rsrctype>article</rsrctype><creationdate>2023</creationdate><recordtype>article</recordtype><sourceid>ABUWG</sourceid><sourceid>AFKRA</sourceid><sourceid>AZQEC</sourceid><sourceid>BENPR</sourceid><sourceid>CCPQU</sourceid><sourceid>DWQXO</sourceid><recordid>eNpjYuA0MjY21LUwMTLiYOAtLs4yMDAwMjM3MjU15mQwd85IzC3IzM9TCM7PKS0BMdLyixRKMlIVwoNdfI0MjIwVQvJz8rMTFcICHRWAqnNyUvPSU3kYWNMSc4pTeaE0N4Oym2uIs4duQVF-YWlqcUl8Vn5pUR5QKt7I3MzC0tDA1MDYmDhVADjEM9k</recordid><startdate>20230220</startdate><enddate>20230220</enddate><creator>Gao, Shengyi</creator><creator>Chen, Zhe</creator><creator>Chen, Guo</creator><creator>Wang, Wenhai</creator><creator>Lu, Tong</creator><general>Cornell University Library, arXiv.org</general><scope>8FE</scope><scope>8FG</scope><scope>ABJCF</scope><scope>ABUWG</scope><scope>AFKRA</scope><scope>AZQEC</scope><scope>BENPR</scope><scope>BGLVJ</scope><scope>CCPQU</scope><scope>DWQXO</scope><scope>HCIFZ</scope><scope>L6V</scope><scope>M7S</scope><scope>PIMPY</scope><scope>PQEST</scope><scope>PQQKQ</scope><scope>PQUKI</scope><scope>PRINS</scope><scope>PTHSS</scope></search><sort><creationdate>20230220</creationdate><title>Champion Solution for the WSDM2023 Toloka VQA Challenge</title><author>Gao, Shengyi ; Chen, Zhe ; Chen, Guo ; Wang, Wenhai ; Lu, Tong</author></sort><facets><frbrtype>5</frbrtype><frbrgroupid>cdi_FETCH-proquest_journals_27689105033</frbrgroupid><rsrctype>articles</rsrctype><prefilter>articles</prefilter><language>eng</language><creationdate>2023</creationdate><topic>Adapters</topic><topic>Questions</topic><topic>Task complexity</topic><topic>Visual tasks</topic><toplevel>online_resources</toplevel><creatorcontrib>Gao, Shengyi</creatorcontrib><creatorcontrib>Chen, Zhe</creatorcontrib><creatorcontrib>Chen, Guo</creatorcontrib><creatorcontrib>Wang, Wenhai</creatorcontrib><creatorcontrib>Lu, Tong</creatorcontrib><collection>ProQuest SciTech Collection</collection><collection>ProQuest Technology Collection</collection><collection>Materials Science &amp; Engineering Collection</collection><collection>ProQuest Central (Alumni Edition)</collection><collection>ProQuest Central UK/Ireland</collection><collection>ProQuest Central Essentials</collection><collection>ProQuest Central</collection><collection>Technology Collection</collection><collection>ProQuest One Community College</collection><collection>ProQuest Central Korea</collection><collection>SciTech Premium Collection</collection><collection>ProQuest Engineering Collection</collection><collection>Engineering Database</collection><collection>Access via ProQuest (Open Access)</collection><collection>ProQuest One Academic Eastern Edition (DO NOT USE)</collection><collection>ProQuest One Academic</collection><collection>ProQuest One Academic UKI Edition</collection><collection>ProQuest Central China</collection><collection>Engineering Collection</collection></facets><delivery><delcategory>Remote Search Resource</delcategory><fulltext>fulltext</fulltext></delivery><addata><au>Gao, Shengyi</au><au>Chen, Zhe</au><au>Chen, Guo</au><au>Wang, Wenhai</au><au>Lu, Tong</au><format>book</format><genre>document</genre><ristype>GEN</ristype><atitle>Champion Solution for the WSDM2023 Toloka VQA Challenge</atitle><jtitle>arXiv.org</jtitle><date>2023-02-20</date><risdate>2023</risdate><eissn>2331-8422</eissn><abstract>In this report, we present our champion solution to the WSDM2023 Toloka Visual Question Answering (VQA) Challenge. Different from the common VQA and visual grounding (VG) tasks, this challenge involves a more complex scenario, i.e. inferring and locating the object implicitly specified by the given interrogative question. For this task, we leverage ViT-Adapter, a pre-training-free adapter network, to adapt multi-modal pre-trained Uni-Perceiver for better cross-modal localization. Our method ranks first on the leaderboard, achieving 77.5 and 76.347 IoU on public and private test sets, respectively. It shows that ViT-Adapter is also an effective paradigm for adapting the unified perception model to vision-language downstream tasks. Code and models will be released at https://github.com/czczup/ViT-Adapter/tree/main/wsdm2023.</abstract><cop>Ithaca</cop><pub>Cornell University Library, arXiv.org</pub><oa>free_for_read</oa></addata></record>
fulltext fulltext
identifier EISSN: 2331-8422
ispartof arXiv.org, 2023-02
issn 2331-8422
language eng
recordid cdi_proquest_journals_2768910503
source Free E- Journals
subjects Adapters
Questions
Task complexity
Visual tasks
title Champion Solution for the WSDM2023 Toloka VQA Challenge
url https://sfx.bib-bvb.de/sfx_tum?ctx_ver=Z39.88-2004&ctx_enc=info:ofi/enc:UTF-8&ctx_tim=2025-01-02T15%3A53%3A03IST&url_ver=Z39.88-2004&url_ctx_fmt=infofi/fmt:kev:mtx:ctx&rfr_id=info:sid/primo.exlibrisgroup.com:primo3-Article-proquest&rft_val_fmt=info:ofi/fmt:kev:mtx:book&rft.genre=document&rft.atitle=Champion%20Solution%20for%20the%20WSDM2023%20Toloka%20VQA%20Challenge&rft.jtitle=arXiv.org&rft.au=Gao,%20Shengyi&rft.date=2023-02-20&rft.eissn=2331-8422&rft_id=info:doi/&rft_dat=%3Cproquest%3E2768910503%3C/proquest%3E%3Curl%3E%3C/url%3E&disable_directlink=true&sfx.directlink=off&sfx.report_link=0&rft_id=info:oai/&rft_pqid=2768910503&rft_id=info:pmid/&rfr_iscdi=true