{"id":"https://openalex.org/W4386902900","doi":"https://doi.org/10.1109/tcsvt.2023.3317447","title":"Video Question Answering With Semantic Disentanglement and Reasoning","display_name":"Video Question Answering With Semantic Disentanglement and Reasoning","publication_year":2023,"publication_date":"2023-09-20","ids":{"openalex":"https://openalex.org/W4386902900","doi":"https://doi.org/10.1109/tcsvt.2023.3317447"},"language":"en","primary_location":{"id":"doi:10.1109/tcsvt.2023.3317447","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsvt.2023.3317447","pdf_url":null,"source":{"id":"https://openalex.org/S115173108","display_name":"IEEE Transactions on Circuits and Systems for Video Technology","issn_l":"1051-8215","issn":["1051-8215","1558-2205"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Circuits and Systems for Video Technology","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5101597404","display_name":"Jin Liu","orcid":"https://orcid.org/0000-0002-5143-2263"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jin Liu","raw_affiliation_strings":["School of Control Science and Engineering, Shandong University, Jinan, China"],"raw_orcid":"https://orcid.org/0000-0002-5143-2263","affiliations":[{"raw_affiliation_string":"School of Control Science and Engineering, Shandong University, Jinan, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100762884","display_name":"Guoxiang Wang","orcid":"https://orcid.org/0000-0002-7467-7466"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guoxiang Wang","raw_affiliation_strings":["School of Control Science and Engineering, Shandong University, Jinan, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Control Science and Engineering, Shandong University, Jinan, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114138234","display_name":"Jialong Xie","orcid":null},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jialong Xie","raw_affiliation_strings":["School of Control Science and Engineering, Shandong University, Jinan, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Control Science and Engineering, Shandong University, Jinan, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085317250","display_name":"Fengyu Zhou","orcid":"https://orcid.org/0000-0001-5140-7036"},"institutions":[{"id":"https://openalex.org/I154099455","display_name":"Shandong University","ror":"https://ror.org/0207yh398","country_code":"CN","type":"education","lineage":["https://openalex.org/I154099455"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fengyu Zhou","raw_affiliation_strings":["School of Control Science and Engineering, Shandong University, Jinan, China"],"raw_orcid":"https://orcid.org/0000-0001-5140-7036","affiliations":[{"raw_affiliation_string":"School of Control Science and Engineering, Shandong University, Jinan, China","institution_ids":["https://openalex.org/I154099455"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5021080053","display_name":"Huijuan Xu","orcid":"https://orcid.org/0000-0002-1858-1822"},"institutions":[{"id":"https://openalex.org/I130769515","display_name":"Pennsylvania State University","ror":"https://ror.org/04p491231","country_code":"US","type":"education","lineage":["https://openalex.org/I130769515"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Huijuan Xu","raw_affiliation_strings":["School of Electrical Engineering and Computer Science, The Pennsylvania State University, University Park, PA, USA"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Electrical Engineering and Computer Science, The Pennsylvania State University, University Park, PA, USA","institution_ids":["https://openalex.org/I130769515"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.4493,"has_fulltext":false,"cited_by_count":25,"citation_normalized_percentile":{"value":0.9060789,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":98,"max":99},"biblio":{"volume":"34","issue":"5","first_page":"3663","last_page":"3673"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":1.0,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9948999881744385,"subfield":{"id":"https://openalex.org/subfields/1702","display_name":"Artificial Intelligence"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.98580002784729,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"field":{"id":"https://openalex.org/fields/17","display_name":"Computer Science"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.8647881150245667},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.6832582950592041},{"id":"https://openalex.org/keywords/question-answering","display_name":"Question answering","score":0.6825156211853027},{"id":"https://openalex.org/keywords/sentence","display_name":"Sentence","score":0.5750460028648376},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.5463923811912537},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.5405884981155396},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.4639911651611328},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.4503242075443268},{"id":"https://openalex.org/keywords/noun-phrase","display_name":"Noun phrase","score":0.4479840397834778},{"id":"https://openalex.org/keywords/comprehension","display_name":"Comprehension","score":0.4343799948692322},{"id":"https://openalex.org/keywords/noun","display_name":"Noun","score":0.40406301617622375}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8647881150245667},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6832582950592041},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.6825156211853027},{"id":"https://openalex.org/C2777530160","wikidata":"https://www.wikidata.org/wiki/Q41796","display_name":"Sentence","level":2,"score":0.5750460028648376},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.5463923811912537},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5405884981155396},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.4639911651611328},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.4503242075443268},{"id":"https://openalex.org/C153962237","wikidata":"https://www.wikidata.org/wiki/Q1401131","display_name":"Noun phrase","level":3,"score":0.4479840397834778},{"id":"https://openalex.org/C511192102","wikidata":"https://www.wikidata.org/wiki/Q5156948","display_name":"Comprehension","level":2,"score":0.4343799948692322},{"id":"https://openalex.org/C121934690","wikidata":"https://www.wikidata.org/wiki/Q1084","display_name":"Noun","level":2,"score":0.40406301617622375},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcsvt.2023.3317447","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsvt.2023.3317447","pdf_url":null,"source":{"id":"https://openalex.org/S115173108","display_name":"IEEE Transactions on Circuits and Systems for Video Technology","issn_l":"1051-8215","issn":["1051-8215","1558-2205"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Circuits and Systems for Video Technology","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.6700000166893005,"id":"https://metadata.un.org/sdg/4"}],"awards":[{"id":"https://openalex.org/G539818514","display_name":null,"funder_award_id":"2019IRS19","funder_id":"https://openalex.org/F4320335791","funder_display_name":"Beijing Advanced Innovation Center for Intelligent Robots and Systems, Beijing Institute of Technology"},{"id":"https://openalex.org/G5486054263","display_name":null,"funder_award_id":"ZR2019MF064","funder_id":"https://openalex.org/F4320324174","funder_display_name":"Natural Science Foundation of Shandong Province"}],"funders":[{"id":"https://openalex.org/F4320324174","display_name":"Natural Science Foundation of Shandong Province","ror":null},{"id":"https://openalex.org/F4320335791","display_name":"Beijing Advanced Innovation Center for Intelligent Robots and Systems, Beijing Institute of Technology","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":73,"referenced_works":["https://openalex.org/W1933349210","https://openalex.org/W2064675550","https://openalex.org/W2194775991","https://openalex.org/W2250539671","https://openalex.org/W2277195237","https://openalex.org/W2425121537","https://openalex.org/W2549139847","https://openalex.org/W2606982687","https://openalex.org/W2745461083","https://openalex.org/W2765716052","https://openalex.org/W2896457183","https://openalex.org/W2899663614","https://openalex.org/W2904452845","https://openalex.org/W2954199749","https://openalex.org/W2962949233","https://openalex.org/W2970476646","https://openalex.org/W2972073579","https://openalex.org/W2981582341","https://openalex.org/W2993371911","https://openalex.org/W2997344006","https://openalex.org/W2997805943","https://openalex.org/W2998166190","https://openalex.org/W3010714856","https://openalex.org/W3027790991","https://openalex.org/W3034730770","https://openalex.org/W3043840704","https://openalex.org/W3119243803","https://openalex.org/W3126942607","https://openalex.org/W3172863135","https://openalex.org/W3173367591","https://openalex.org/W3175961224","https://openalex.org/W3176796859","https://openalex.org/W3179621320","https://openalex.org/W3189890868","https://openalex.org/W3197457832","https://openalex.org/W3202778561","https://openalex.org/W3203995540","https://openalex.org/W3204670646","https://openalex.org/W3205789812","https://openalex.org/W3206675006","https://openalex.org/W3209229003","https://openalex.org/W4200631219","https://openalex.org/W4205492448","https://openalex.org/W4221142658","https://openalex.org/W4224917597","https://openalex.org/W4224926131","https://openalex.org/W4226060302","https://openalex.org/W4283750241","https://openalex.org/W4283801188","https://openalex.org/W4283814553","https://openalex.org/W4285191490","https://openalex.org/W4285600797","https://openalex.org/W4306294758","https://openalex.org/W4312246181","https://openalex.org/W4312974690","https://openalex.org/W4313071966","https://openalex.org/W4313175608","https://openalex.org/W4377235294","https://openalex.org/W4383112678","https://openalex.org/W4385245566","https://openalex.org/W4385572108","https://openalex.org/W4386071468","https://openalex.org/W6755207826","https://openalex.org/W6755977528","https://openalex.org/W6759579507","https://openalex.org/W6774818996","https://openalex.org/W6780226713","https://openalex.org/W6797109355","https://openalex.org/W6801948084","https://openalex.org/W6803578320","https://openalex.org/W6804277322","https://openalex.org/W6810488671","https://openalex.org/W6846313647"],"related_works":["https://openalex.org/W2384605597","https://openalex.org/W2387743295","https://openalex.org/W3082787378","https://openalex.org/W2136007095","https://openalex.org/W2366230879","https://openalex.org/W3208425359","https://openalex.org/W2349927912","https://openalex.org/W3159777597","https://openalex.org/W1976369278","https://openalex.org/W2025590371"],"abstract_inverted_index":{"Video":[0],"question":[1,128,170],"answering":[2],"aims":[3],"to":[4,61,72,82,109,147,179,187],"provide":[5],"correct":[6],"answers":[7,155],"given":[8],"complex":[9],"videos":[10],"and":[11,24,53,89,95,122,154,167,193,203,208],"related":[12],"questions,":[13],"posting":[14],"high":[15],"requirements":[16],"of":[17,51,76,87,106,211],"the":[18,39,43,47,57,63,73,77,84,92,111,126,149,157,206,221],"comprehension":[19],"ability":[20],"in":[21,56,70],"both":[22,165],"video":[23,40,97,169],"language":[25,104,177],"processing.":[26],"Existing":[27],"works":[28],"phrase":[29],"this":[30,68],"task":[31],"as":[32,144,183],"a":[33,136,175],"multi-modal":[34,58],"fusion":[35],"process":[36,60],"by":[37,140],"aligning":[38],"context":[41,118,186],"with":[42,117],"whole":[44,78],"question,":[45],"ignoring":[46],"rich":[48],"semantic":[49,74,85,112],"details":[50],"nouns":[52],"verbs":[54],"separately":[55],"reasoning":[59],"derive":[62],"final":[64],"answer.":[65],"To":[66,124],"fill":[67],"gap,":[69],"addition":[71],"alignment":[75],"sentence,":[79],"we":[80,133],"propose":[81],"disentangle":[83],"understanding":[86,113,129],"language,":[88],"reason":[90],"over":[91],"corresponding":[93],"frame-level":[94],"motion-level":[96],"features.":[98],"We":[99,172],"design":[100,135],"an":[101],"unified":[102],"multi-granularity":[103],"module":[105],"residual":[107],"structure":[108],"adapt":[110],"at":[114],"different":[115],"granularity":[116],"exchange,":[119],"e.g.,":[120],"word-level":[121],"sentence-level.":[123],"enhance":[125],"holistic":[127],"for":[130,164],"answer":[131,185],"prediction,":[132],"also":[134],"contrastive":[137],"sampling":[138],"approach":[139],"selecting":[141],"irrelevant":[142],"questions":[143,153],"negative":[145],"samples":[146],"break":[148],"intrinsic":[150],"correlations":[151],"between":[152],"within":[156],"dataset.":[158],"Notably,":[159],"our":[160,212],"model":[161,178],"is":[162],"competent":[163],"multiple-choice":[166],"open-ended":[168,189],"answering.":[171],"further":[173],"employ":[174],"pre-trained":[176],"retrieve":[180],"relevant":[181],"knowledge":[182],"candidate":[184],"facilitate":[188],"VideoQA.":[190],"Extensive":[191],"quantitative":[192],"qualitative":[194],"experiments":[195],"on":[196],"four":[197],"public":[198],"datasets":[199],"(NextQA,":[200],"MSVD,":[201],"MSRVTT,":[202],"TGIF-QA-R)":[204],"demonstrate":[205],"effective":[207],"superior":[209],"performance":[210],"proposed":[213],"model.":[214],"Our":[215],"code":[216],"will":[217],"be":[218],"released":[219],"upon":[220],"paper\u2019s":[222],"acceptance.":[223]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":12},{"year":2024,"cited_by_count":10}],"updated_date":"2026-07-18T07:39:51.176621","created_date":"2025-10-10T00:00:00"}