{"id":"https://openalex.org/W4306874801","doi":"https://doi.org/10.1145/3568679","title":"Hierarchical Multi-Attention Transfer for Knowledge Distillation","display_name":"Hierarchical Multi-Attention Transfer for Knowledge Distillation","publication_year":2022,"publication_date":"2022-10-20","ids":{"openalex":"https://openalex.org/W4306874801","doi":"https://doi.org/10.1145/3568679"},"language":"en","primary_location":{"id":"doi:10.1145/3568679","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3568679","pdf_url":null,"source":{"id":"https://openalex.org/S19610489","display_name":"ACM Transactions on Multimedia Computing Communications and Applications","issn_l":"1551-6857","issn":["1551-6857","1551-6865"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Multimedia Computing, Communications, and Applications","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/A5042645332","display_name":"Jianping Gou","orcid":"https://orcid.org/0000-0003-1413-0693"},"institutions":[{"id":"https://openalex.org/I115592961","display_name":"Jiangsu University","ror":"https://ror.org/03jc41j30","country_code":"CN","type":"education","lineage":["https://openalex.org/I115592961"]},{"id":"https://openalex.org/I142108993","display_name":"Southwest University","ror":"https://ror.org/01kj4z117","country_code":"CN","type":"education","lineage":["https://openalex.org/I142108993"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jianping Gou","raw_affiliation_strings":["College of Computer and Information Science, College of Software, Southwest University, China School of Computer Science and Communication Engineering, Jiangsu University, China"],"raw_orcid":"https://orcid.org/0000-0003-1413-0693","affiliations":[{"raw_affiliation_string":"College of Computer and Information Science, College of Software, Southwest University, China School of Computer Science and Communication Engineering, Jiangsu University, China","institution_ids":["https://openalex.org/I115592961","https://openalex.org/I142108993"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048524847","display_name":"Liyuan Sun","orcid":"https://orcid.org/0000-0002-6079-5541"},"institutions":[{"id":"https://openalex.org/I115592961","display_name":"Jiangsu University","ror":"https://ror.org/03jc41j30","country_code":"CN","type":"education","lineage":["https://openalex.org/I115592961"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liyuan Sun","raw_affiliation_strings":["School of Computer Science and Communication Engineering, Jiangsu University, China"],"raw_orcid":"https://orcid.org/0000-0002-6079-5541","affiliations":[{"raw_affiliation_string":"School of Computer Science and Communication Engineering, Jiangsu University, China","institution_ids":["https://openalex.org/I115592961"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085309099","display_name":"Baosheng Yu","orcid":"https://orcid.org/0000-0002-0761-7893"},"institutions":[{"id":"https://openalex.org/I129604602","display_name":"The University of Sydney","ror":"https://ror.org/0384j8v12","country_code":"AU","type":"education","lineage":["https://openalex.org/I129604602"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Baosheng Yu","raw_affiliation_strings":["School of Computer Science, The University of Sydney, Australia"],"raw_orcid":"https://orcid.org/0000-0002-0761-7893","affiliations":[{"raw_affiliation_string":"School of Computer Science, The University of Sydney, Australia","institution_ids":["https://openalex.org/I129604602"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003203521","display_name":"Shaohua Wan","orcid":"https://orcid.org/0000-0001-7013-9081"},"institutions":[{"id":"https://openalex.org/I150229711","display_name":"University of Electronic Science and Technology of China","ror":"https://ror.org/04qr3zq92","country_code":"CN","type":"education","lineage":["https://openalex.org/I150229711"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shaohua Wan","raw_affiliation_strings":["Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China, China"],"raw_orcid":"https://orcid.org/0000-0001-7013-9081","affiliations":[{"raw_affiliation_string":"Shenzhen Institute for Advanced Study, University of Electronic Science and Technology of China, China","institution_ids":["https://openalex.org/I150229711"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5074103823","display_name":"Dacheng Tao","orcid":"https://orcid.org/0000-0001-7225-5449"},"institutions":[{"id":"https://openalex.org/I129604602","display_name":"The University of Sydney","ror":"https://ror.org/0384j8v12","country_code":"AU","type":"education","lineage":["https://openalex.org/I129604602"]}],"countries":["AU"],"is_corresponding":false,"raw_author_name":"Dacheng Tao","raw_affiliation_strings":["School of Computer Science, The University of Sydney, Australia"],"raw_orcid":"https://orcid.org/0000-0001-7225-5449","affiliations":[{"raw_affiliation_string":"School of Computer Science, The University of Sydney, Australia","institution_ids":["https://openalex.org/I129604602"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":6.0738,"has_fulltext":false,"cited_by_count":79,"citation_normalized_percentile":{"value":0.97707889,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":89,"max":100},"biblio":{"volume":"20","issue":"2","first_page":"1","last_page":"20"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network 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/T10036","display_name":"Advanced Neural Network 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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.9998000264167786,"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/T11714","display_name":"Multimodal Machine Learning Applications","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/computer-science","display_name":"Computer science","score":0.8657952547073364},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.6832178235054016},{"id":"https://openalex.org/keywords/knowledge-transfer","display_name":"Knowledge transfer","score":0.6509185433387756},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.6172246932983398},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.5610806941986084},{"id":"https://openalex.org/keywords/transfer-of-learning","display_name":"Transfer of learning","score":0.546628475189209},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.5369569659233093},{"id":"https://openalex.org/keywords/deep-learning","display_name":"Deep learning","score":0.49293041229248047},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.48133715987205505},{"id":"https://openalex.org/keywords/knowledge-engineering","display_name":"Knowledge engineering","score":0.43523940443992615},{"id":"https://openalex.org/keywords/artificial-neural-network","display_name":"Artificial neural network","score":0.4163789749145508},{"id":"https://openalex.org/keywords/domain-knowledge","display_name":"Domain knowledge","score":0.41491320729255676},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.35169869661331177},{"id":"https://openalex.org/keywords/knowledge-management","display_name":"Knowledge management","score":0.10054701566696167}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8657952547073364},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6832178235054016},{"id":"https://openalex.org/C2776960227","wikidata":"https://www.wikidata.org/wiki/Q2586354","display_name":"Knowledge transfer","level":2,"score":0.6509185433387756},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6172246932983398},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.5610806941986084},{"id":"https://openalex.org/C150899416","wikidata":"https://www.wikidata.org/wiki/Q1820378","display_name":"Transfer of learning","level":2,"score":0.546628475189209},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.5369569659233093},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.49293041229248047},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.48133715987205505},{"id":"https://openalex.org/C84685590","wikidata":"https://www.wikidata.org/wiki/Q1540472","display_name":"Knowledge engineering","level":2,"score":0.43523940443992615},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.4163789749145508},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.41491320729255676},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.35169869661331177},{"id":"https://openalex.org/C56739046","wikidata":"https://www.wikidata.org/wiki/Q192060","display_name":"Knowledge management","level":1,"score":0.10054701566696167},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C138885662","wikidata":"https://www.wikidata.org/wiki/Q5891","display_name":"Philosophy","level":0,"score":0.0},{"id":"https://openalex.org/C178790620","wikidata":"https://www.wikidata.org/wiki/Q11351","display_name":"Organic chemistry","level":1,"score":0.0},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.0},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.0},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3568679","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3568679","pdf_url":null,"source":{"id":"https://openalex.org/S19610489","display_name":"ACM Transactions on Multimedia Computing Communications and Applications","issn_l":"1551-6857","issn":["1551-6857","1551-6865"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Multimedia Computing, Communications, and Applications","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.8299999833106995}],"awards":[{"id":"https://openalex.org/G3406750037","display_name":null,"funder_award_id":"61976107","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G7519899501","display_name":null,"funder_award_id":"FL-170100117","funder_id":"https://openalex.org/F4320334704","funder_display_name":"Australian Research Council"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320321605","display_name":"Government of Jiangsu Province","ror":"https://ror.org/004svx814"},{"id":"https://openalex.org/F4320327794","display_name":"Qinglan Project of Jiangsu Province of China","ror":null},{"id":"https://openalex.org/F4320334704","display_name":"Australian Research Council","ror":"https://ror.org/05mmh0f86"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":48,"referenced_works":["https://openalex.org/W639708223","https://openalex.org/W1586532344","https://openalex.org/W1690739335","https://openalex.org/W1821462560","https://openalex.org/W1849277567","https://openalex.org/W2108598243","https://openalex.org/W2133564696","https://openalex.org/W2138011018","https://openalex.org/W2190008860","https://openalex.org/W2194775991","https://openalex.org/W2300242332","https://openalex.org/W2550553598","https://openalex.org/W2613718673","https://openalex.org/W2731516819","https://openalex.org/W2739879705","https://openalex.org/W2743289088","https://openalex.org/W2786945063","https://openalex.org/W2883780447","https://openalex.org/W2939308562","https://openalex.org/W2962858109","https://openalex.org/W2962965870","https://openalex.org/W2963026686","https://openalex.org/W2963163009","https://openalex.org/W2963534679","https://openalex.org/W2964137095","https://openalex.org/W2995607862","https://openalex.org/W3008374555","https://openalex.org/W3026860374","https://openalex.org/W3034368386","https://openalex.org/W3046606162","https://openalex.org/W3100877062","https://openalex.org/W3108855944","https://openalex.org/W3136272401","https://openalex.org/W3138154797","https://openalex.org/W3161703143","https://openalex.org/W3173799170","https://openalex.org/W3177052299","https://openalex.org/W4205581758","https://openalex.org/W4205662443","https://openalex.org/W4206210197","https://openalex.org/W4295262505","https://openalex.org/W4297734170","https://openalex.org/W6638523607","https://openalex.org/W6693397755","https://openalex.org/W6713132643","https://openalex.org/W6726275242","https://openalex.org/W6769906912","https://openalex.org/W6785881197"],"related_works":["https://openalex.org/W2600782034","https://openalex.org/W2342033266","https://openalex.org/W1484858421","https://openalex.org/W1538105803","https://openalex.org/W2082812939","https://openalex.org/W1993567425","https://openalex.org/W2046128842","https://openalex.org/W2127091547","https://openalex.org/W129258864","https://openalex.org/W2107532841"],"abstract_inverted_index":{"Knowledge":[0],"distillation":[1,22],"(KD)":[2],"is":[3,23],"a":[4,28,33,91],"powerful":[5],"and":[6,121,130,135,145,161],"widely":[7],"applicable":[8],"technique":[9],"for":[10,51,83,116],"the":[11,38,66,74,107,126,140,166],"compression":[12],"of":[13,20,69,77,101,112],"deep":[14,70,78,113],"learning":[15,80,115],"models.":[16],"The":[17],"main":[18],"idea":[19],"knowledge":[21,26,64,84,108,117,124,127,138,141],"to":[24,32,47,105],"transfer":[25,61,94,106,170],"from":[27,65,128,142],"large":[29],"teacher":[30],"model":[31],"small":[34],"student":[35],"model,":[36],"where":[37,98],"attention":[39,63,102,123,137],"mechanism":[40],"has":[41],"been":[42],"intensively":[43],"explored":[44],"in":[45],"regard":[46],"its":[48],"great":[49],"flexibility":[50],"managing":[52],"different":[53,99,110],"teacher-student":[54],"architectures.":[55],"However,":[56],"existing":[57],"attention-based":[58],"methods":[59],"usually":[60],"similar":[62],"intermediate":[67],"layers":[68],"neural":[71],"networks,":[72],"leaving":[73],"hierarchical":[75,92,168],"structure":[76],"representation":[79,114],"poorly":[81],"investigated":[82],"distillation.":[85,118],"In":[86],"this":[87],"paper,":[88],"we":[89],"propose":[90],"multi-attention":[93,169],"framework":[95],"(HMAT)":[96],",":[97],"types":[100],"are":[103],"utilized":[104],"at":[109],"levels":[111],"Specifically,":[119],"position-based":[120],"channel-based":[122],"characterize":[125,139],"low-level":[129],"high-level":[131,146],"feature":[132,147],"representations,":[133],"respectively,":[134],"activation-based":[136],"both":[143],"mid-level":[144],"representations.":[148],"Extensive":[149],"experiments":[150],"on":[151],"three":[152],"popular":[153],"visual":[154],"recognition":[155],"tasks,":[156],"image":[157,159],"classification,":[158],"retrieval,":[160],"object":[162],"detection,":[163],"demonstrate":[164],"that":[165],"proposed":[167],"or":[171],"HMAT":[172],"significantly":[173],"outperforms":[174],"recent":[175],"state-of-the-art":[176],"KD":[177],"methods.":[178]},"counts_by_year":[{"year":2026,"cited_by_count":4},{"year":2025,"cited_by_count":31},{"year":2024,"cited_by_count":29},{"year":2023,"cited_by_count":14},{"year":2022,"cited_by_count":1}],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-10T00:00:00"}