{"id":"https://openalex.org/W4385767942","doi":"https://doi.org/10.24963/ijcai.2023/104","title":"Text-Video Retrieval with Disentangled Conceptualization and Set-to-Set Alignment","display_name":"Text-Video Retrieval with Disentangled Conceptualization and Set-to-Set Alignment","publication_year":2023,"publication_date":"2023-08-01","ids":{"openalex":"https://openalex.org/W4385767942","doi":"https://doi.org/10.24963/ijcai.2023/104"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2023/104","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2023/104","pdf_url":"https://www.ijcai.org/proceedings/2023/0104.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://www.ijcai.org/proceedings/2023/0104.pdf","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5100731314","display_name":"Peng Jin","orcid":"https://orcid.org/0000-0003-2228-131X"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Peng Jin","raw_affiliation_strings":["AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","School of Electronic and Computer Engineering, Peking University, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Electronic and Computer Engineering, Peking University, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100348632","display_name":"Hao Li","orcid":"https://orcid.org/0000-0002-7712-0890"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Li","raw_affiliation_strings":["AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","School of Electronic and Computer Engineering, Peking University, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Electronic and Computer Engineering, Peking University, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026357570","display_name":"Zesen Cheng","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zesen Cheng","raw_affiliation_strings":["AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","School of Electronic and Computer Engineering, Peking University, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Electronic and Computer Engineering, Peking University, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110947034","display_name":"Jinfa Huang","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinfa Huang","raw_affiliation_strings":["AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","School of Electronic and Computer Engineering, Peking University, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"School of Electronic and Computer Engineering, Peking University, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101874417","display_name":"Zhennan Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhennan Wang","raw_affiliation_strings":["Peng Cheng Laboratory, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Peng Cheng Laboratory, Shenzhen, China","institution_ids":["https://openalex.org/I4210136793"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100409808","display_name":"Yuan Li","orcid":"https://orcid.org/0000-0002-4462-6764"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li Yuan","raw_affiliation_strings":["AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","Peng Cheng Laboratory, Shenzhen, China","School of Electronic and Computer Engineering, Peking University, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"Peng Cheng Laboratory, Shenzhen, China","institution_ids":["https://openalex.org/I4210136793"]},{"raw_affiliation_string":"School of Electronic and Computer Engineering, Peking University, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100353141","display_name":"Chang Liu","orcid":"https://orcid.org/0000-0001-5560-7203"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chang Liu","raw_affiliation_strings":["Department of Automation and BNRist, Tsinghua University, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Department of Automation and BNRist, Tsinghua University, Beijing, China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100332984","display_name":"Jie Chen","orcid":"https://orcid.org/0000-0002-9765-4523"},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]},{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jie Chen","raw_affiliation_strings":["AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","Peng Cheng Laboratory, Shenzhen, China","School of Electronic and Computer Engineering, Peking University, Shenzhen, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"AI for Science (AI4S)-Preferred Program, Peking University Shenzhen Graduate School, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]},{"raw_affiliation_string":"Peng Cheng Laboratory, Shenzhen, China","institution_ids":["https://openalex.org/I4210136793"]},{"raw_affiliation_string":"School of Electronic and Computer Engineering, Peking University, Shenzhen, China","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5100731314"],"corresponding_institution_ids":["https://openalex.org/I20231570"],"apc_list":null,"apc_paid":null,"fwci":3.7564,"has_fulltext":false,"cited_by_count":33,"citation_normalized_percentile":{"value":0.94866848,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":100},"biblio":{"volume":null,"issue":null,"first_page":"938","last_page":"946"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9990000128746033,"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/T11439","display_name":"Video Analysis and Summarization","score":0.9983000159263611,"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/conceptualization","display_name":"Conceptualization","score":0.8482012152671814},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7998595833778381},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.6302074193954468},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.5640729069709778},{"id":"https://openalex.org/keywords/pooling","display_name":"Pooling","score":0.4937969148159027},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.46589767932891846},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.42360424995422363},{"id":"https://openalex.org/keywords/semantic-matching","display_name":"Semantic matching","score":0.4133238196372986},{"id":"https://openalex.org/keywords/theoretical-computer-science","display_name":"Theoretical computer science","score":0.39356279373168945},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.38279712200164795},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.3771268427371979},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.3752601146697998},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3502451181411743},{"id":"https://openalex.org/keywords/mathematics","display_name":"Mathematics","score":0.08056235313415527}],"concepts":[{"id":"https://openalex.org/C90734943","wikidata":"https://www.wikidata.org/wiki/Q17008777","display_name":"Conceptualization","level":2,"score":0.8482012152671814},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7998595833778381},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.6302074193954468},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.5640729069709778},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.4937969148159027},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.46589767932891846},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.42360424995422363},{"id":"https://openalex.org/C2778493491","wikidata":"https://www.wikidata.org/wiki/Q7449072","display_name":"Semantic matching","level":3,"score":0.4133238196372986},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.39356279373168945},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.38279712200164795},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3771268427371979},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3752601146697998},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3502451181411743},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.08056235313415527},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","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/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2023/104","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2023/104","pdf_url":"https://www.ijcai.org/proceedings/2023/0104.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.24963/ijcai.2023/104","is_oa":true,"landing_page_url":"https://doi.org/10.24963/ijcai.2023/104","pdf_url":"https://www.ijcai.org/proceedings/2023/0104.pdf","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Second International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","score":0.7200000286102295,"display_name":"Quality Education"}],"awards":[{"id":"https://openalex.org/G3229855046","display_name":null,"funder_award_id":"61972217","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G3262489766","display_name":null,"funder_award_id":"62176249","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G4768617711","display_name":null,"funder_award_id":"32071459","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5052683820","display_name":null,"funder_award_id":"2019B1515120049","funder_id":"https://openalex.org/F4320321921","funder_display_name":"Natural Science Foundation of Guangdong Province"},{"id":"https://openalex.org/G5598344535","display_name":null,"funder_award_id":"62006133","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7571408944","display_name":null,"funder_award_id":"62271465","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320321921","display_name":"Natural Science Foundation of Guangdong Province","ror":null}],"has_content":{"grobid_xml":false,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4385767942.pdf"},"referenced_works_count":49,"referenced_works":["https://openalex.org/W1522301498","https://openalex.org/W2078238240","https://openalex.org/W2164290393","https://openalex.org/W2183868163","https://openalex.org/W2425121537","https://openalex.org/W2737047298","https://openalex.org/W2786385136","https://openalex.org/W2942718970","https://openalex.org/W2945420892","https://openalex.org/W2951428711","https://openalex.org/W2963017553","https://openalex.org/W2963226019","https://openalex.org/W2963916161","https://openalex.org/W2965458216","https://openalex.org/W2970210011","https://openalex.org/W2971127577","https://openalex.org/W2972073579","https://openalex.org/W3035356601","https://openalex.org/W3043840704","https://openalex.org/W3094502228","https://openalex.org/W3122640483","https://openalex.org/W3153005511","https://openalex.org/W3157480928","https://openalex.org/W3166396011","https://openalex.org/W3168640669","https://openalex.org/W3174873881","https://openalex.org/W3175759175","https://openalex.org/W3176799298","https://openalex.org/W3204588463","https://openalex.org/W3211549568","https://openalex.org/W4221146248","https://openalex.org/W4280632620","https://openalex.org/W4280635869","https://openalex.org/W4285606530","https://openalex.org/W4295990387","https://openalex.org/W4297808394","https://openalex.org/W4297899894","https://openalex.org/W4309804076","https://openalex.org/W4312299780","https://openalex.org/W4312372711","https://openalex.org/W4312734855","https://openalex.org/W4312999114","https://openalex.org/W4313196315","https://openalex.org/W4327485959","https://openalex.org/W4382142048","https://openalex.org/W4382202590","https://openalex.org/W4385245566","https://openalex.org/W4386076265","https://openalex.org/W4390889802"],"related_works":["https://openalex.org/W4362575170","https://openalex.org/W1532772865","https://openalex.org/W2531253623","https://openalex.org/W2984230149","https://openalex.org/W2112095637","https://openalex.org/W2237121679","https://openalex.org/W4390616414","https://openalex.org/W2006867289","https://openalex.org/W3212890177","https://openalex.org/W1188194610"],"abstract_inverted_index":{"Text-video":[0],"retrieval":[1],"is":[2,123],"a":[3,83,90,119,133],"challenging":[4],"cross-modal":[5],"task,":[6],"which":[7],"aims":[8],"to":[9,21,34,53,76,89,101,105],"align":[10],"visual":[11,86],"entities":[12],"with":[13],"natural":[14],"language":[15],"descriptions.":[16],"Current":[17],"methods":[18],"either":[19],"fail":[20,33],"leverage":[22,35],"the":[23,36,46,55,68,107,157],"local":[24],"details":[25],"or":[26],"are":[27],"computationally":[28],"expensive.":[29],"What's":[30],"worse,":[31],"they":[32],"heterogeneous":[37],"concepts":[38,87,104,115],"in":[39,117],"data.":[40],"In":[41,110],"this":[42],"paper,":[43],"we":[44,66,95,113],"propose":[45,96],"Disentangled":[47],"Conceptualization":[48],"and":[49,57,127,150],"Set-to-set":[50],"Alignment":[51],"(DiCoSA)":[52],"simulate":[54],"conceptualizing":[56],"reasoning":[58],"process":[59],"of":[60,85,92],"human":[61],"beings.":[62],"For":[63,79],"disentangled":[64],"conceptualization,":[65],"divide":[67],"coarse":[69],"feature":[70],"into":[71],"multiple":[72],"latent":[73],"factors":[74],"related":[75],"semantic":[77,103],"concepts.":[78],"set-to-set":[80],"alignment,":[81],"where":[82],"set":[84,91],"correspond":[88],"textual":[93],"concepts,":[94],"an":[97],"adaptive":[98],"pooling":[99],"method":[100,155],"aggregate":[102],"address":[106],"partial":[108],"matching.":[109],"particular,":[111],"since":[112],"encode":[114],"independently":[116],"only":[118],"few":[120],"dimensions,":[121],"DiCoSA":[122],"superior":[124],"at":[125],"efficiency":[126],"granularity,":[128],"ensuring":[129],"fine-grained":[130],"interactions":[131],"using":[132],"similar":[134],"computational":[135],"complexity":[136],"as":[137],"coarse-grained":[138],"alignment.":[139],"Extensive":[140],"experiments":[141],"on":[142],"five":[143],"datasets,":[144],"including":[145],"MSR-VTT,":[146],"LSMDC,":[147],"MSVD,":[148],"ActivityNet,":[149],"DiDeMo,":[151],"demonstrate":[152],"that":[153],"our":[154],"outperforms":[156],"existing":[158],"state-of-the-art":[159],"methods.":[160]},"counts_by_year":[{"year":2026,"cited_by_count":5},{"year":2025,"cited_by_count":11},{"year":2024,"cited_by_count":13},{"year":2023,"cited_by_count":4}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}