{"id":"https://openalex.org/W4415708719","doi":"https://doi.org/10.1109/icme59968.2025.11210112","title":"Bridging the One-to-Many Gap: Multi-label Semantic Learning and Relay for Video Captioning","display_name":"Bridging the One-to-Many Gap: Multi-label Semantic Learning and Relay for Video Captioning","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W4415708719","doi":"https://doi.org/10.1109/icme59968.2025.11210112"},"language":null,"primary_location":{"id":"doi:10.1109/icme59968.2025.11210112","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icme59968.2025.11210112","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","raw_type":"proceedings-article"},"type":"conference-paper","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/A5101736130","display_name":"Shuqin Chen","orcid":"https://orcid.org/0000-0003-3436-653X"},"institutions":[{"id":"https://openalex.org/I4210154851","display_name":"Hubei University of Education","ror":"https://ror.org/04f41cb79","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210154851"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shuqin Chen","raw_affiliation_strings":["Hubei University of Education,School of Computer"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hubei University of Education,School of Computer","institution_ids":["https://openalex.org/I4210154851"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101237483","display_name":"Yikang Hu","orcid":null},"institutions":[{"id":"https://openalex.org/I4210165606","display_name":"Hubei Normal University","ror":"https://ror.org/056y3dw16","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210165606"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yikang Hu","raw_affiliation_strings":["Hubei Normal University,College of Computer and Information Engineering"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hubei Normal University,College of Computer and Information Engineering","institution_ids":["https://openalex.org/I4210165606"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086759035","display_name":"Li Yang","orcid":"https://orcid.org/0000-0003-3821-9342"},"institutions":[{"id":"https://openalex.org/I4210154851","display_name":"Hubei University of Education","ror":"https://ror.org/04f41cb79","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210154851"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li Yang","raw_affiliation_strings":["Hubei University of Education,School of Computer"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hubei University of Education,School of Computer","institution_ids":["https://openalex.org/I4210154851"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101456842","display_name":"Zhixin Sun","orcid":"https://orcid.org/0000-0001-6148-9688"},"institutions":[{"id":"https://openalex.org/I4210120238","display_name":"PowerChina (China)","ror":"https://ror.org/01varr368","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210120238"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhixin Sun","raw_affiliation_strings":["PowerChina Zhongnan Engineering Corporation Limited"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"PowerChina Zhongnan Engineering Corporation Limited","institution_ids":["https://openalex.org/I4210120238"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5055585044","display_name":"Liangjun Yu","orcid":"https://orcid.org/0000-0002-1324-0500"},"institutions":[{"id":"https://openalex.org/I4210154851","display_name":"Hubei University of Education","ror":"https://ror.org/04f41cb79","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210154851"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liangjun Yu","raw_affiliation_strings":["Hubei University of Education,School of Computer"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Hubei University of Education,School of Computer","institution_ids":["https://openalex.org/I4210154851"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5064738683","display_name":"Xian Zhong","orcid":"https://orcid.org/0000-0002-5242-0467"},"institutions":[{"id":"https://openalex.org/I196699116","display_name":"Wuhan University of Technology","ror":"https://ror.org/03fe7t173","country_code":"CN","type":"education","lineage":["https://openalex.org/I196699116"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xian Zhong","raw_affiliation_strings":["Wuhan University of Technology,Hubei Key Laboratory of Transportation Internet of Things"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Wuhan University of Technology,Hubei Key Laboratory of Transportation Internet of Things","institution_ids":["https://openalex.org/I196699116"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9674999713897705,"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":0.9674999713897705,"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":0.007300000172108412,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.0031999999191612005,"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/closed-captioning","display_name":"Closed captioning","score":0.8288999795913696},{"id":"https://openalex.org/keywords/bridging","display_name":"Bridging (networking)","score":0.7782999873161316},{"id":"https://openalex.org/keywords/ambiguity","display_name":"Ambiguity","score":0.6031000018119812},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.5859000086784363},{"id":"https://openalex.org/keywords/granularity","display_name":"Granularity","score":0.4763000011444092},{"id":"https://openalex.org/keywords/relay","display_name":"Relay","score":0.42899999022483826}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8385999798774719},{"id":"https://openalex.org/C157657479","wikidata":"https://www.wikidata.org/wiki/Q2367247","display_name":"Closed captioning","level":3,"score":0.8288999795913696},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.7782999873161316},{"id":"https://openalex.org/C2780522230","wikidata":"https://www.wikidata.org/wiki/Q1140419","display_name":"Ambiguity","level":2,"score":0.6031000018119812},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.5859000086784363},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48170000314712524},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.4763000011444092},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.43779999017715454},{"id":"https://openalex.org/C2778156585","wikidata":"https://www.wikidata.org/wiki/Q174053","display_name":"Relay","level":3,"score":0.42899999022483826},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3677999973297119},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.3276999890804291},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.30959999561309814},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.3084999918937683},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2858000099658966},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2632000148296356},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.25589999556541443}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icme59968.2025.11210112","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icme59968.2025.11210112","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322186","display_name":"Natural Science Foundation of Hubei Province","ror":null},{"id":"https://openalex.org/F4320324848","display_name":"Hubei University of Education","ror":"https://ror.org/04f41cb79"},{"id":"https://openalex.org/F4320337504","display_name":"Research and Development","ror":"https://ror.org/027s68j25"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":26,"referenced_works":["https://openalex.org/W1956340063","https://openalex.org/W2101105183","https://openalex.org/W2425121537","https://openalex.org/W2526050071","https://openalex.org/W2593116425","https://openalex.org/W2962934715","https://openalex.org/W2970641574","https://openalex.org/W2989322838","https://openalex.org/W3101167253","https://openalex.org/W3134875898","https://openalex.org/W3176689360","https://openalex.org/W3207461616","https://openalex.org/W4220790454","https://openalex.org/W4281560470","https://openalex.org/W4284692156","https://openalex.org/W4293518015","https://openalex.org/W4372260221","https://openalex.org/W4372266864","https://openalex.org/W4382464395","https://openalex.org/W4384161812","https://openalex.org/W4386057717","https://openalex.org/W4386122702","https://openalex.org/W4386159762","https://openalex.org/W4387934928","https://openalex.org/W4400269657","https://openalex.org/W4400850904"],"related_works":[],"abstract_inverted_index":{"Many":[0],"commonly":[1],"used":[2],"video":[3],"captioning":[4],"datasets":[5],"contain":[6],"multiple":[7,79],"caption":[8],"annotations":[9],"per":[10],"video.":[11],"When":[12],"training":[13],"with":[14],"cross-entropy":[15],"loss,":[16],"the":[17,22,39,49,67,115,119],"model":[18],"encounters":[19],"ambiguity":[20],"because":[21],"same":[23],"input":[24],"is":[25,139],"mapped":[26],"to":[27,31,85,109,128],"different":[28],"targets,":[29],"leading":[30],"confusion.":[32],"To":[33],"address":[34],"this":[35],"issue,":[36],"we":[37],"propose":[38],"Multi-label":[40,70,99],"Semantic":[41,100],"Learning":[42,72],"and":[43,54,81],"Relay":[44,101],"(MSLR)":[45],"framework,":[46],"which":[47],"transforms":[48],"one-to-many":[50],"relation":[51],"between":[52],"videos":[53],"their":[55],"descriptions":[56,80,124],"into":[57,114],"a":[58,106],"one-to-one":[59],"mapping.":[60],"Specifically,":[61],"MSLR":[62,132],"introduces":[63],"two":[64],"modules":[65],"after":[66],"decoder.":[68],"The":[69,98,137],"Multi-granularity":[71],"(MML)":[73],"module":[74,103],"integrates":[75],"sentence-level":[76],"granularity":[77,88],"from":[78,95],"employs":[82],"attention":[83],"mechanisms":[84],"extract":[86],"word-level":[87],"weights,":[89],"thereby":[90],"capturing":[91],"complementary":[92,111],"semantic":[93],"information":[94],"diverse":[96],"perspectives.":[97],"(MSR)":[102],"subsequently":[104],"leverages":[105],"parameter-sharing":[107],"mechanism":[108],"feed":[110],"semantics":[112],"back":[113],"decoder,":[116],"thus":[117],"preventing":[118],"generation":[120],"of":[121],"overly":[122],"generic":[123],"during":[125],"inference.":[126],"Compared":[127],"state-of-the-art":[129],"lightweight":[130],"methods,":[131],"achieves":[133],"highly":[134],"competitive":[135],"results.":[136],"code":[138],"available":[140],"at":[141],"https://github.com/hyk0320/MSLR.":[142]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-30T00:00:00"}