{"id":"https://openalex.org/W4403420955","doi":"https://doi.org/10.1109/tcsvt.2024.3480949","title":"Selectively Hard Negative Mining for Alleviating Gradient Vanishing in Image-Text Matching","display_name":"Selectively Hard Negative Mining for Alleviating Gradient Vanishing in Image-Text Matching","publication_year":2024,"publication_date":"2024-10-15","ids":{"openalex":"https://openalex.org/W4403420955","doi":"https://doi.org/10.1109/tcsvt.2024.3480949"},"language":"en","primary_location":{"id":"doi:10.1109/tcsvt.2024.3480949","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsvt.2024.3480949","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/A5100415053","display_name":"Zheng Li","orcid":"https://orcid.org/0000-0003-2535-2523"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zheng Li","raw_affiliation_strings":["School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-2535-2523","affiliations":[{"raw_affiliation_string":"School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058648594","display_name":"Caili Guo","orcid":"https://orcid.org/0000-0001-8892-4520"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Caili Guo","raw_affiliation_strings":["School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-8892-4520","affiliations":[{"raw_affiliation_string":"School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102949527","display_name":"Xin Wang","orcid":"https://orcid.org/0009-0006-6463-4338"},"institutions":[{"id":"https://openalex.org/I139759216","display_name":"Beijing University of Posts and Telecommunications","ror":"https://ror.org/04w9fbh59","country_code":"CN","type":"education","lineage":["https://openalex.org/I139759216"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xin Wang","raw_affiliation_strings":["School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0006-6463-4338","affiliations":[{"raw_affiliation_string":"School of Information and Communication Engineering, Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5019172074","display_name":"Zerun Feng","orcid":"https://orcid.org/0000-0003-3987-0591"},"institutions":[{"id":"https://openalex.org/I4210100255","display_name":"Beijing Academy of Artificial Intelligence","ror":"https://ror.org/016a74861","country_code":"CN","type":"other","lineage":["https://openalex.org/I4210100255"]},{"id":"https://openalex.org/I4210136246","display_name":"China Telecom (China)","ror":"https://ror.org/03jgnzt20","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210136246"]},{"id":"https://openalex.org/I4387153335","display_name":"China Telecom","ror":"https://ror.org/05p67dv18","country_code":null,"type":"company","lineage":["https://openalex.org/I4387153335"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zerun Feng","raw_affiliation_strings":["Institute of Artificial Intelligence (TeleAI), China, Telecom, Beijing, China","Institute of Artificial Intelligence (TeleAI), China Telecom, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0003-3987-0591","affiliations":[{"raw_affiliation_string":"Institute of Artificial Intelligence (TeleAI), China, Telecom, Beijing, China","institution_ids":["https://openalex.org/I4210100255"]},{"raw_affiliation_string":"Institute of Artificial Intelligence (TeleAI), China Telecom, Beijing, China","institution_ids":["https://openalex.org/I4210136246","https://openalex.org/I4387153335"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5110965933","display_name":"Zhongtian Du","orcid":null},"institutions":[{"id":"https://openalex.org/I4210136246","display_name":"China Telecom (China)","ror":"https://ror.org/03jgnzt20","country_code":"CN","type":"company","lineage":["https://openalex.org/I4210136246"]},{"id":"https://openalex.org/I4387153335","display_name":"China Telecom","ror":"https://ror.org/05p67dv18","country_code":null,"type":"company","lineage":["https://openalex.org/I4387153335"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhongtian Du","raw_affiliation_strings":["China Telecom Digital Intelligence Technology Company Ltd., Beijing, China","China Telecom Digital Intelligence Technology Co., Ltd, Beijing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"China Telecom Digital Intelligence Technology Company Ltd., Beijing, China","institution_ids":["https://openalex.org/I4210136246"]},{"raw_affiliation_string":"China Telecom Digital Intelligence Technology Co., Ltd, Beijing, China","institution_ids":["https://openalex.org/I4210136246","https://openalex.org/I4387153335"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":2.1873,"has_fulltext":false,"cited_by_count":10,"citation_normalized_percentile":{"value":0.89173487,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":99},"biblio":{"volume":"35","issue":"2","first_page":"1921","last_page":"1935"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.9879000186920166,"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/T10824","display_name":"Image Retrieval and Classification Techniques","score":0.9879000186920166,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.9785000085830688,"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/T10601","display_name":"Handwritten Text Recognition Techniques","score":0.9322999715805054,"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":0.5457645058631897},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.5345247387886047},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4754469096660614},{"id":"https://openalex.org/keywords/image-processing","display_name":"Image processing","score":0.43746885657310486},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.4278087019920349},{"id":"https://openalex.org/keywords/image-segmentation","display_name":"Image segmentation","score":0.41402748227119446},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.406170129776001}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5457645058631897},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.5345247387886047},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4754469096660614},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.43746885657310486},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4278087019920349},{"id":"https://openalex.org/C124504099","wikidata":"https://www.wikidata.org/wiki/Q56933","display_name":"Image segmentation","level":3,"score":0.41402748227119446},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.406170129776001}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcsvt.2024.3480949","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsvt.2024.3480949","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":[],"awards":[{"id":"https://openalex.org/G2914057848","display_name":null,"funder_award_id":"2021XD-A01-1","funder_id":"https://openalex.org/F4320335787","funder_display_name":"Fundamental Research Funds for the Central Universities"},{"id":"https://openalex.org/G5448899510","display_name":null,"funder_award_id":"62371070","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8982324736","display_name":null,"funder_award_id":"2021XD-A01-1","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/F4320335787","display_name":"Fundamental Research Funds for the Central Universities","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":63,"referenced_works":["https://openalex.org/W1861492603","https://openalex.org/W2096733369","https://openalex.org/W2105482032","https://openalex.org/W2138011018","https://openalex.org/W2138621090","https://openalex.org/W2185175083","https://openalex.org/W2194775991","https://openalex.org/W2340690086","https://openalex.org/W2745461083","https://openalex.org/W2883311563","https://openalex.org/W2896457183","https://openalex.org/W2950635152","https://openalex.org/W2962964995","https://openalex.org/W2963026686","https://openalex.org/W2965848243","https://openalex.org/W2982078236","https://openalex.org/W2988823324","https://openalex.org/W2994818707","https://openalex.org/W3005971801","https://openalex.org/W3010277541","https://openalex.org/W3035212740","https://openalex.org/W3035356601","https://openalex.org/W3035454331","https://openalex.org/W3035552787","https://openalex.org/W3035588244","https://openalex.org/W3035605030","https://openalex.org/W3088460296","https://openalex.org/W3094172275","https://openalex.org/W3108373531","https://openalex.org/W3109894131","https://openalex.org/W3110042533","https://openalex.org/W3118694826","https://openalex.org/W3131251978","https://openalex.org/W3155230099","https://openalex.org/W3169472988","https://openalex.org/W3175888430","https://openalex.org/W3177343494","https://openalex.org/W3206597437","https://openalex.org/W4206314411","https://openalex.org/W4210894218","https://openalex.org/W4224952037","https://openalex.org/W4225166070","https://openalex.org/W4226391640","https://openalex.org/W4282968790","https://openalex.org/W4283812943","https://openalex.org/W4285118104","https://openalex.org/W4312341105","https://openalex.org/W4312463400","https://openalex.org/W4312761738","https://openalex.org/W4312940158","https://openalex.org/W4312989333","https://openalex.org/W4313181088","https://openalex.org/W4322576838","https://openalex.org/W4323797084","https://openalex.org/W4403481522","https://openalex.org/W6600609147","https://openalex.org/W6620707391","https://openalex.org/W6636510571","https://openalex.org/W6678470764","https://openalex.org/W6747225742","https://openalex.org/W6757817989","https://openalex.org/W6774481907","https://openalex.org/W6803448863"],"related_works":["https://openalex.org/W2755342338","https://openalex.org/W2779427294","https://openalex.org/W2775347418","https://openalex.org/W2625805835","https://openalex.org/W2079911747","https://openalex.org/W3116076068","https://openalex.org/W3003936178","https://openalex.org/W2145652935","https://openalex.org/W2563206327","https://openalex.org/W2069885731"],"abstract_inverted_index":{"Most":[0],"Image-Text":[1],"Matching":[2],"(ITM)":[3],"models":[4,38,160],"adopt":[5],"Triplet":[6,131],"loss":[7,132],"with":[8,133,185],"Hard":[9,135],"Negative":[10,136],"mining":[11,137],"(T-HN)":[12],"as":[13],"the":[14,19,44,68,76,88,104,108,116,144,150,173,208],"optimization":[15],"objective.":[16],"T-HN":[17],"mines":[18],"hardest":[20,89,145],"negative":[21,90,146],"samples":[22,91,147],"in":[23,43,161],"each":[24],"batch":[25],"for":[26],"training":[27,41,50,72,168,197],"and":[28,55,195,198],"achieves":[29],"impressive":[30],"performance.":[31,214],"However,":[32],"we":[33,65,101,128,177],"observe":[34],"that":[35,67,75],"these":[36],"ITM":[37,84,159,205],"have":[39],"bad":[40,71],"behaviors":[42],"early":[45],"phases":[46],"of":[47,70,175,210],"training.":[48,112],"Model":[49],"is":[51,58,74,217],"difficult":[52],"to":[53,60,95,119,142,149,158,165],"converge,":[54],"matching":[56],"performance":[57],"slow":[59],"improve.":[61],"In":[62],"this":[63],"paper,":[64],"find":[66],"cause":[69],"behavior":[73],"model":[77,85,184],"suffers":[78],"from":[79],"gradient":[80,96,99,109,117,126,151],"vanishing.":[81,97],"Optimizing":[82],"an":[83],"using":[86],"only":[87],"can":[92,155],"easily":[93],"lead":[94],"Through":[98],"analysis,":[100],"first":[102],"derive":[103],"condition":[105],"under":[106,121],"which":[107,139],"vanishes":[110],"during":[111],"We":[113],"explain":[114],"why":[115],"tends":[118],"zero":[120],"certain":[122],"conditions.":[123],"To":[124,170],"alleviate":[125],"vanishing,":[127],"propose":[129],"a":[130,162,179,191],"Selectively":[134],"(T-SelHN),":[138],"decides":[140],"whether":[141],"mine":[143],"according":[148],"vanishing":[152],"condition.":[153],"T-SelHN":[154],"be":[156],"applied":[157],"plug-and-play":[163],"manner":[164],"improve":[166],"their":[167],"behaviors.":[169],"further":[171],"ensure":[172],"back-propagation":[174],"gradients,":[176],"construct":[178],"Residual":[180],"Visual":[181],"Semantic":[182],"Embedding":[183],"T-SelHN,":[186],"denoted":[187],"RVSE++,":[188,211],"which\u00c2":[189],"has":[190],"simple":[192],"network":[193],"structure":[194],"efficient":[196],"inference":[199],"speeds.":[200],"Extensive":[201],"experiments":[202],"on":[203],"two":[204],"benchmarks":[206],"demonstrate":[207],"strength":[209],"achieving":[212],"state-of-the-art":[213],"The":[215],"code":[216],"available":[218],"at":[219],"<uri":[220],"xmlns:mml=\"http://www.w3.org/1998/Math/MathML\"":[221],"xmlns:xlink=\"http://www.w3.org/1999/xlink\">https://github.com/AAA-Zheng/RVSEPP</uri>.":[222]},"counts_by_year":[{"year":2025,"cited_by_count":9},{"year":2024,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}