{"id":"https://openalex.org/W4415541331","doi":"https://doi.org/10.1145/3746027.3754844","title":"Dynamic Self-adaptive Multiscale Distillation from Pre-trained Multimodal Large Model for Efficient Cross-modal Retrieval","display_name":"Dynamic Self-adaptive Multiscale Distillation from Pre-trained Multimodal Large Model for Efficient Cross-modal Retrieval","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415541331","doi":"https://doi.org/10.1145/3746027.3754844"},"language":null,"primary_location":{"id":"doi:10.1145/3746027.3754844","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3754844","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd 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":null,"display_name":"Zhengyang Liang","orcid":"https://orcid.org/0009-0008-0205-0163"},"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":true,"raw_author_name":"Zhengyang Liang","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China and Singapore Management University, Singapore, Singapore"],"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China and Singapore Management University, Singapore, Singapore","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056818951","display_name":"Meiyu Liang","orcid":"https://orcid.org/0000-0003-1835-1848"},"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":"Meiyu Liang","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5113868566","display_name":"Wei Po Huang","orcid":"https://orcid.org/0009-0007-1661-9501"},"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":"Wei Huang","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100633553","display_name":"Yawen Li","orcid":"https://orcid.org/0000-0003-2662-3444"},"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":"Yawen Li","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068917997","display_name":"Wu Liu","orcid":"https://orcid.org/0000-0003-1633-7575"},"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":"Wu Liu","raw_affiliation_strings":["University of Science and Technology of China, Hefei, China"],"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/A5014615052","display_name":"Yingxia Shao","orcid":"https://orcid.org/0000-0002-8559-2628"},"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":"Yingxia Shao","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5034044616","display_name":"Kangkang Lu","orcid":"https://orcid.org/0000-0003-0983-758X"},"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":"Kangkang Lu","raw_affiliation_strings":["Beijing University of Posts and Telecommunications, Beijing, China"],"affiliations":[{"raw_affiliation_string":"Beijing University of Posts and Telecommunications, Beijing, China","institution_ids":["https://openalex.org/I139759216"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I139759216"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.16042645,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"2851","last_page":"2859"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.996999979019165,"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"}},"topics":[{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.996999979019165,"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/T10627","display_name":"Advanced Image and Video Retrieval Techniques","score":0.996399998664856,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9957000017166138,"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/distillation","display_name":"Distillation","score":0.6740000247955322},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4092999994754791},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.39719998836517334},{"id":"https://openalex.org/keywords/software-deployment","display_name":"Software deployment","score":0.35409998893737793},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.3068000078201294},{"id":"https://openalex.org/keywords/calibration","display_name":"Calibration","score":0.2653000056743622}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8183000087738037},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6740000247955322},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5085999965667725},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5031999945640564},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4092999994754791},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.39719998836517334},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.35409998893737793},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3068000078201294},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2703999876976013},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.2653000056743622},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.2651999890804291},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.26019999384880066},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.25699999928474426}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3746027.3754844","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3754844","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G377032498","display_name":null,"funder_award_id":"2023ZD0121503","funder_id":"https://openalex.org/F4320329860","funder_display_name":"National Science and Technology Major Project"},{"id":"https://openalex.org/G5820838707","display_name":null,"funder_award_id":"No. 20230484319","funder_id":"https://openalex.org/F4320334978","funder_display_name":"Beijing Nova Program"},{"id":"https://openalex.org/G6407883800","display_name":null,"funder_award_id":"No. U23A20319, 62192784, U24A20329, 62272054, 62422202","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/F4320329860","display_name":"National Science and Technology Major Project","ror":null},{"id":"https://openalex.org/F4320334978","display_name":"Beijing Nova Program","ror":"https://ror.org/034k14f91"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":20,"referenced_works":["https://openalex.org/W2117539524","https://openalex.org/W2739879705","https://openalex.org/W2962964995","https://openalex.org/W2963140444","https://openalex.org/W2964111476","https://openalex.org/W2982078236","https://openalex.org/W2982157312","https://openalex.org/W2988823324","https://openalex.org/W3155230099","https://openalex.org/W3175888430","https://openalex.org/W4226426325","https://openalex.org/W4304092117","https://openalex.org/W4312877428","https://openalex.org/W4365801687","https://openalex.org/W4366241040","https://openalex.org/W4386071498","https://openalex.org/W4386071687","https://openalex.org/W4386075677","https://openalex.org/W4391305547","https://openalex.org/W4393160343"],"related_works":[],"abstract_inverted_index":{"In":[0,92],"recent":[1],"years,":[2],"pre-trained":[3,49,108],"multimodal":[4,18,50],"large":[5,51,110],"models":[6,52],"have":[7],"attracted":[8],"widespread":[9],"attention":[10],"due":[11],"to":[12,47,80,143],"their":[13,30],"outstanding":[14],"performance":[15,74,178,210,224],"in":[16,37,159],"various":[17],"applications.":[19],"Nonetheless,":[20],"the":[21,85,98,121,141,166,192,207],"extensive":[22],"computational":[23,41],"resources":[24],"and":[25,128,162],"vast":[26],"datasets":[27],"required":[28],"for":[29,35,97,112],"training":[31,167],"present":[32],"significant":[33,73,177],"hurdles":[34],"deployment":[36],"environments":[38],"with":[39],"limited":[40],"resources.":[42],"Many":[43],"existing":[44,230],"methods":[45,65],"attempt":[46],"compress":[48],"through":[53],"knowledge":[54,88],"distillation,":[55],"typically":[56],"focusing":[57],"on":[58,225],"a":[59,101,136,160],"single":[60],"optimization":[61,78,197],"objective.":[62],"While":[63],"such":[64],"successfully":[66],"reduce":[67],"model":[68,111,203,220],"parameters,":[69],"they":[70],"often":[71],"incur":[72],"degradation.":[75],"Moreover,":[76],"single-scale":[77,182],"fails":[79],"ensure":[81],"comprehensive":[82],"learning":[83],"of":[84,123,206,215],"teacher":[86,208],"model's":[87,209],"across":[89,198],"different":[90],"aspects.":[91],"this":[93],"work,":[94],"we":[95,134],"propose,":[96],"first":[99],"time,":[100],"dynamic":[102,137,151,188],"self-adaptive":[103],"multiscale":[104,173],"distillation":[105,174,183,193],"(DSMD)":[106],"from":[107,120],"multi-modal":[109],"efficient":[113],"cross-modal":[114,226],"retrieval":[115,227],"method,":[116],"considering":[117],"multiple":[118],"scales":[119],"perspectives":[122],"fine":[124],"granularity,":[125],"global":[126],"structure,":[127],"hard":[129],"negative":[130],"sample":[131],"mining.":[132],"Furthermore,":[133],"design":[135],"loss":[138],"balancer,":[139],"eliminating":[140],"need":[142],"manually":[144],"tune":[145],"objective":[146],"weights":[147],"during":[148],"distillation.":[149],"This":[150],"mechanism":[152],"ensures":[153],"that":[154,171],"all":[155],"objectives":[156],"are":[157,233],"optimized":[158],"balanced":[161],"adaptive":[163],"manner":[164],"throughout":[165],"process.":[168],"Experiments":[169],"demonstrate":[170],"our":[172,186,219],"framework":[175],"achieves":[176,204,222],"improvements":[179],"over":[180],"traditional":[181],"methods.":[184],"Additionally,":[185],"proposed":[187],"balancer":[189],"effectively":[190],"stabilizes":[191],"process,":[194],"ensuring":[195],"consistent":[196],"objectives.":[199],"The":[200],"distilled":[201],"student":[202],"90%":[205],"while":[211],"using":[212],"only":[213],"10%":[214],"its":[216],"parameters.":[217],"Notably,":[218],"also":[221],"state-of-the-art":[223],"tasks,":[228],"outperforming":[229],"approaches.":[231],"Codes":[232],"available":[234],"at":[235],"https://github.com/chrisx599/DSMD.":[236]},"counts_by_year":[],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-25T00:00:00"}