{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,12]],"date-time":"2026-07-12T04:52:35Z","timestamp":1783831955336,"version":"3.55.0"},"reference-count":69,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["2238208"],"award-info":[{"award-number":["2238208"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["DGE-1922591"],"award-info":[{"award-number":["DGE-1922591"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["1909912"],"award-info":[{"award-number":["1909912"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["DMR-2202124"],"award-info":[{"award-number":["DMR-2202124"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006168","name":"National Nuclear Security Administration","doi-asserted-by":"publisher","award":["NA0004078"],"award-info":[{"award-number":["NA0004078"]}],"id":[{"id":"10.13039\/100006168","id-type":"DOI","asserted-by":"publisher"}]},{"name":"NIH","award":["R01EY034562"],"award-info":[{"award-number":["R01EY034562"]}]},{"DOI":"10.13039\/100000185","name":"Defense Advanced Research Projects Agency","doi-asserted-by":"publisher","award":["HR00112220003"],"award-info":[{"award-number":["HR00112220003"]}],"id":[{"id":"10.13039\/100000185","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Center of Excellence in Data Science"},{"name":"Cisco"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Multimedia"],"published-print":{"date-parts":[[2024]]},"DOI":"10.1109\/tmm.2024.3380259","type":"journal-article","created":{"date-parts":[[2024,3,21]],"date-time":"2024-03-21T18:30:49Z","timestamp":1711045849000},"page":"8609-8624","source":"Crossref","is-referenced-by-count":43,"title":["Cross Modality Bias in Visual Question Answering: A Causal View With Possible Worlds VQA"],"prefix":"10.1109","volume":"26","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-1014-2937","authenticated-orcid":false,"given":"Ali","family":"Vosoughi","sequence":"first","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of Rochester, Rochester, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-9560-702X","authenticated-orcid":false,"given":"Shijian","family":"Deng","sequence":"additional","affiliation":[{"name":"Department of Computer Science, University of Texas Dallas, Dallas, TX, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4316-3320","authenticated-orcid":false,"given":"Songyang","family":"Zhang","sequence":"additional","affiliation":[{"name":"Department of Computer Science, University of Rochester, Rochester, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1423-4513","authenticated-orcid":false,"given":"Yapeng","family":"Tian","sequence":"additional","affiliation":[{"name":"Department of Computer Science, University of Texas Dallas, Dallas, TX, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2183-822X","authenticated-orcid":false,"given":"Chenliang","family":"Xu","sequence":"additional","affiliation":[{"name":"Department of Computer Science, University of Rochester, Rochester, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4516-9729","authenticated-orcid":false,"given":"Jiebo","family":"Luo","sequence":"additional","affiliation":[{"name":"Department of Computer Science, University of Rochester, Rochester, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"16292","article-title":"Introspective distillation for robust question answering","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Niu","year":"2021"},{"key":"ref2","first-page":"841","article-title":"RUBi: Reducing unimodal biases for visual question answering","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Cadene","year":"2019"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00209"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i07.6776"},{"key":"ref5","first-page":"1541","article-title":"Overcoming language priors in visual question answering with adversarial regularization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ramakrishnan","year":"2018"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-1418"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01251"},{"key":"ref8","first-page":"3197","article-title":"Removing bias in multi-modal classifiers: Regularization by maximizing functional entropies","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Gat","year":"2020"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1016\/j.tics.2020.08.005"},{"key":"ref10","author":"Rubin","year":"1915","journal-title":"Synsoplevede Figurer"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-016-0966-6"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00522"},{"key":"ref13","first-page":"2591","article-title":"Taking a HINT: Leveraging explanations to make vision and language models more grounded","volume-title":"Proc. IEEE\/CVF Int. Conf. Comput. Vis.","author":"Ramprasaath","year":"2019"},{"key":"ref14","first-page":"8604","article-title":"Self-critical reasoning for robust visual question answering","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wu","year":"2019"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1016\/j.cviu.2017.10.001"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01081"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01006"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00145"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2020\/151"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/WACV51458.2022.00263"},{"key":"ref21","first-page":"2376","article-title":"Counterfactual visual explanations","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Goyal","year":"2019"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00900"},{"key":"ref23","first-page":"1","article-title":"CLEVRER: Collision events for video representation and reasoning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Yi","year":"2020"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00377"},{"key":"ref25","first-page":"2734","article-title":"Interventional few-shot learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Yue","year":"2020"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01515"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00395"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01077"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/3394171.3413518"},{"key":"ref30","first-page":"655","article-title":"Causal intervention for weakly-supervised semantic segmentation","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhang","year":"2020"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58607-2_34"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3121705"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.357"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00972"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.818"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.586"},{"key":"ref37","volume-title":"The Book of Why: The New Science of Cause and Effect","author":"Pearl","year":"2018"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1203"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.542"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.670"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-acl.432"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3292597"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.573"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2022.3225754"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00089"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2023.3305817"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00502"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01124"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2020.3026892"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00504"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01830"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2023.3302162"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01216-8_17"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01922"},{"key":"ref55","volume-title":"Visuell Wahrgenommene Figuren: Studien in Psychologischer Analyse","volume":"1","author":"Rubin","year":"1921"},{"key":"ref56","first-page":"1255","article-title":"On causal and anticausal learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Janzing","year":"2012"},{"key":"ref57","article-title":"Invariant risk minimization","author":"Arjovsky","year":"2019"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1145\/3501714.3501736"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.10"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00636"},{"key":"ref61","first-page":"12888","article-title":"BLIP: Bootstrapping language-image pre-training for unified vision-language understanding and generation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Li","year":"2022"},{"key":"ref62","first-page":"19730","article-title":"BLIP-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Li","year":"2023"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58601-0_2"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00161"},{"key":"ref65","first-page":"407","article-title":"On the value of out-of-distribution testing: An example of Goodharts law","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Teney","year":"2020"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.265"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.63"},{"key":"ref68","article-title":"Distilling the knowledge in a neural network","author":"Hinton","year":"2015"},{"key":"ref69","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Radford","year":"2021"}],"container-title":["IEEE Transactions on Multimedia"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/6046\/10384483\/10477535-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/6046\/10384483\/10477535.pdf?arnumber=10477535","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,8,30]],"date-time":"2024-08-30T10:24:37Z","timestamp":1725013477000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10477535\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024]]},"references-count":69,"URL":"https:\/\/doi.org\/10.1109\/tmm.2024.3380259","relation":{},"ISSN":["1520-9210","1941-0077"],"issn-type":[{"value":"1520-9210","type":"print"},{"value":"1941-0077","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024]]}}}