{"id":"https://openalex.org/W4388187937","doi":"https://doi.org/10.1145/3581783.3612537","title":"Learning Semantics-Grounded Vocabulary Representation for Video-Text Retrieval","display_name":"Learning Semantics-Grounded Vocabulary Representation for Video-Text Retrieval","publication_year":2023,"publication_date":"2023-10-26","ids":{"openalex":"https://openalex.org/W4388187937","doi":"https://doi.org/10.1145/3581783.3612537"},"language":"en","primary_location":{"id":"doi:10.1145/3581783.3612537","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3581783.3612537","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM International Conference on Multimedia","raw_type":"proceedings-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/A5019915855","display_name":"Yaya Shi","orcid":"https://orcid.org/0000-0003-0465-6712"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yaya Shi","raw_affiliation_strings":["University of Science and Technology of China, Hefei, China"],"raw_orcid":"https://orcid.org/0000-0003-0465-6712","affiliations":[{"raw_affiliation_string":"University of Science and Technology of China, Hefei, China","institution_ids":["https://openalex.org/I126520041"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075202003","display_name":"Haowei Liu","orcid":"https://orcid.org/0000-0002-0439-2692"},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haowei Liu","raw_affiliation_strings":["Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-0439-2692","affiliations":[{"raw_affiliation_string":"Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210112150","https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066544410","display_name":"Haiyang Xu","orcid":"https://orcid.org/0000-0001-9442-5912"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Haiyang Xu","raw_affiliation_strings":["DAMO Academy, Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0001-9442-5912","affiliations":[{"raw_affiliation_string":"DAMO Academy, Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008327536","display_name":"Zongyang Ma","orcid":"https://orcid.org/0009-0003-1980-9379"},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zongyang Ma","raw_affiliation_strings":["Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0003-1980-9379","affiliations":[{"raw_affiliation_string":"Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210112150","https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005965903","display_name":"Qinghao Ye","orcid":"https://orcid.org/0000-0002-7977-5540"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Qinghao Ye","raw_affiliation_strings":["DAMO Academy, Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-7977-5540","affiliations":[{"raw_affiliation_string":"DAMO Academy, Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019498452","display_name":"Anwen Hu","orcid":"https://orcid.org/0000-0001-8839-4996"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Anwen Hu","raw_affiliation_strings":["DAMO Academy, Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0001-8839-4996","affiliations":[{"raw_affiliation_string":"DAMO Academy, Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100452477","display_name":"Ming Yan","orcid":"https://orcid.org/0000-0003-4959-8878"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ming Yan","raw_affiliation_strings":["DAMO Academy, Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0003-4959-8878","affiliations":[{"raw_affiliation_string":"DAMO Academy, Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100329266","display_name":"Ji Zhang","orcid":"https://orcid.org/0000-0002-3835-7975"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ji Zhang","raw_affiliation_strings":["DAMO Academy, Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-3835-7975","affiliations":[{"raw_affiliation_string":"DAMO Academy, Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101488344","display_name":"Fei Huang","orcid":"https://orcid.org/0000-0002-3709-5053"},"institutions":[{"id":"https://openalex.org/I45928872","display_name":"Alibaba Group (China)","ror":"https://ror.org/00k642b80","country_code":"CN","type":"company","lineage":["https://openalex.org/I45928872"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fei Huang","raw_affiliation_strings":["DAMO Academy, Alibaba Group, Hangzhou, China"],"raw_orcid":"https://orcid.org/0000-0002-3709-5053","affiliations":[{"raw_affiliation_string":"DAMO Academy, Alibaba Group, Hangzhou, China","institution_ids":["https://openalex.org/I45928872"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5005377211","display_name":"Chunfeng Yuan","orcid":"https://orcid.org/0000-0003-2219-4961"},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chunfeng Yuan","raw_affiliation_strings":["Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-2219-4961","affiliations":[{"raw_affiliation_string":"Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210112150","https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100451204","display_name":"Bing Li","orcid":"https://orcid.org/0000-0001-6114-1411"},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bing Li","raw_affiliation_strings":["Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-6114-1411","affiliations":[{"raw_affiliation_string":"Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210112150","https://openalex.org/I4210165038"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114549594","display_name":"Weiming Hu","orcid":"https://orcid.org/0000-0001-9237-8825"},"institutions":[{"id":"https://openalex.org/I4210112150","display_name":"Institute of Automation","ror":"https://ror.org/022c3hy66","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210112150"]},{"id":"https://openalex.org/I4210165038","display_name":"University of Chinese Academy of Sciences","ror":"https://ror.org/05qbk4x57","country_code":"CN","type":"education","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210165038"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weiming Hu","raw_affiliation_strings":["Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-9237-8825","affiliations":[{"raw_affiliation_string":"Institute of Automation, CAS & University of Chinese Academy of Sciences, Beijing, China","institution_ids":["https://openalex.org/I4210112150","https://openalex.org/I4210165038"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5003217535","display_name":"Zheng-Jun Zha","orcid":"https://orcid.org/0000-0003-2510-8993"},"institutions":[{"id":"https://openalex.org/I126520041","display_name":"University of Science and Technology of China","ror":"https://ror.org/04c4dkn09","country_code":"CN","type":"education","lineage":["https://openalex.org/I126520041","https://openalex.org/I19820366"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng-Jun Zha","raw_affiliation_strings":["University of Science and Technology of China, Hefei, China"],"raw_orcid":"https://orcid.org/0000-0003-2510-8993","affiliations":[{"raw_affiliation_string":"University of Science and Technology of China, Hefei, China","institution_ids":["https://openalex.org/I126520041"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":13,"corresponding_author_ids":["https://openalex.org/A5019915855"],"corresponding_institution_ids":["https://openalex.org/I126520041"],"apc_list":null,"apc_paid":null,"fwci":0.3532,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.60451306,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"4460","last_page":"4470"},"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9995999932289124,"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":0.9933000206947327,"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":0.780478835105896},{"id":"https://openalex.org/keywords/natural-language-processing","display_name":"Natural language processing","score":0.6779283285140991},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.6584134101867676},{"id":"https://openalex.org/keywords/vocabulary","display_name":"Vocabulary","score":0.6311188340187073},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6086896657943726},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4824448227882385},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.4697806239128113},{"id":"https://openalex.org/keywords/modality","display_name":"Modality (human\u2013computer interaction)","score":0.4588220417499542},{"id":"https://openalex.org/keywords/dimension","display_name":"Dimension (graph theory)","score":0.4292908310890198},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.39928609132766724},{"id":"https://openalex.org/keywords/linguistics","display_name":"Linguistics","score":0.15347471833229065},{"id":"https://openalex.org/keywords/programming-language","display_name":"Programming language","score":0.08916768431663513}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.780478835105896},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.6779283285140991},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.6584134101867676},{"id":"https://openalex.org/C2777601683","wikidata":"https://www.wikidata.org/wiki/Q6499736","display_name":"Vocabulary","level":2,"score":0.6311188340187073},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6086896657943726},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4824448227882385},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.4697806239128113},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.4588220417499542},{"id":"https://openalex.org/C33676613","wikidata":"https://www.wikidata.org/wiki/Q13415176","display_name":"Dimension (graph theory)","level":2,"score":0.4292908310890198},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.39928609132766724},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.15347471833229065},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.08916768431663513},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"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/C202444582","wikidata":"https://www.wikidata.org/wiki/Q837863","display_name":"Pure mathematics","level":1,"score":0.0},{"id":"https://openalex.org/C94625758","wikidata":"https://www.wikidata.org/wiki/Q7163","display_name":"Politics","level":2,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0},{"id":"https://openalex.org/C17744445","wikidata":"https://www.wikidata.org/wiki/Q36442","display_name":"Political science","level":0,"score":0.0},{"id":"https://openalex.org/C199539241","wikidata":"https://www.wikidata.org/wiki/Q7748","display_name":"Law","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3581783.3612537","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3581783.3612537","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Quality Education","score":0.7799999713897705,"id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W2034593090","https://openalex.org/W2078238240","https://openalex.org/W2126579184","https://openalex.org/W2963017553","https://openalex.org/W2984008963","https://openalex.org/W2987401211","https://openalex.org/W3035118106","https://openalex.org/W3035372819","https://openalex.org/W3035635319","https://openalex.org/W3126721948","https://openalex.org/W3154280800","https://openalex.org/W3168640669","https://openalex.org/W3204588463","https://openalex.org/W3214192224","https://openalex.org/W3217102353","https://openalex.org/W3217340782","https://openalex.org/W4225432580","https://openalex.org/W4225832925","https://openalex.org/W4287775237","https://openalex.org/W4309804076","https://openalex.org/W4319049530","https://openalex.org/W4367367040","https://openalex.org/W4385574358","https://openalex.org/W6600100092","https://openalex.org/W6600376255"],"related_works":["https://openalex.org/W2385859805","https://openalex.org/W2530972254","https://openalex.org/W4390516098","https://openalex.org/W2374013449","https://openalex.org/W2181948922","https://openalex.org/W73545470","https://openalex.org/W2384362569","https://openalex.org/W2349784553","https://openalex.org/W3022596247","https://openalex.org/W2142795561"],"abstract_inverted_index":{"Previous":[0],"dual-encoder":[1],"pre-training":[2],"methods":[3,148],"for":[4,10],"video-text":[5,42,78,151],"retrieval":[6,152],"employ":[7],"contrastive":[8],"learning":[9],"cross-modal":[11],"alignment":[12],"in":[13,24,44,48,104],"a":[14,34,45,54,59,115,122],"latent":[15,20],"space.":[16,111],"However,":[17],"such":[18],"learned":[19],"spaces":[21],"often":[22],"result":[23],"modality":[25,94,141],"gap":[26,95,142],"problem":[27],"[26].":[28],"In":[29],"this":[30],"paper,":[31],"we":[32,113],"introduce":[33],"novel":[35],"SemVTR":[36,132,144],"framework":[37],"designed":[38],"to":[39,53,88],"learn":[40],"semantics-grounded":[41],"representations":[43,85],"vocabulary":[46,84,110,135],"space,":[47],"which":[49],"each":[50],"dimension":[51],"corresponds":[52],"semantic":[55],"concept":[56],"represented":[57],"by":[58,65],"word.":[60],"The":[61,128],"representation":[62,136],"is":[63],"obtained":[64],"grounding":[66,105],"video":[67,106],"and":[68,91,107,121,137],"text":[69,108],"into":[70,109],"semantically-related":[71],"dimensions":[72],"with":[73],"high":[74],"activation":[75],"values.":[76],"As":[77],"pairs":[79],"share":[80],"grounded":[81],"dimensions,":[82],"their":[83],"are":[86],"expected":[87],"cluster":[89],"together":[90],"thus":[92],"alleviate":[93],"problem.":[96,143],"So,":[97],"the":[98,140],"crux":[99],"of":[100],"our":[101],"method":[102],"lies":[103],"Specifically,":[112],"propose":[114],"Multi-Granularity":[116],"Video":[117],"Semantics":[118,124],"Grounding":[119],"approach":[120],"Textual":[123],"Preserving":[125],"training":[126],"strategy.":[127],"visualization":[129],"illustrates":[130],"that":[131],"obtains":[133],"semantics-gronded":[134],"also":[138],"alleviates":[139],"significantly":[145],"outperforms":[146],"existing":[147],"on":[149],"four":[150],"benchmarks.":[153]},"counts_by_year":[{"year":2024,"cited_by_count":3}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}