{"id":"https://openalex.org/W4408387851","doi":"https://doi.org/10.1007/s11263-025-02374-x","title":"Exemplar-Free Continual Learning of Vision Transformers via Gated Class-Attention and Cascaded Feature Drift Compensation","display_name":"Exemplar-Free Continual Learning of Vision Transformers via Gated Class-Attention and Cascaded Feature Drift Compensation","publication_year":2025,"publication_date":"2025-03-13","ids":{"openalex":"https://openalex.org/W4408387851","doi":"https://doi.org/10.1007/s11263-025-02374-x"},"language":"en","primary_location":{"id":"doi:10.1007/s11263-025-02374-x","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11263-025-02374-x","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11263-025-02374-x.pdf","source":{"id":"https://openalex.org/S25538012","display_name":"International Journal of Computer Vision","issn_l":"0920-5691","issn":["0920-5691","1573-1405"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Journal of Computer Vision","raw_type":"journal-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://link.springer.com/content/pdf/10.1007/s11263-025-02374-x.pdf","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5030534371","display_name":"Marco Cotogni","orcid":"https://orcid.org/0000-0001-7950-7370"},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":true,"raw_author_name":"Marco Cotogni","raw_affiliation_strings":["Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy"],"affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy","institution_ids":["https://openalex.org/I25217355"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100365435","display_name":"Fei Yang","orcid":"https://orcid.org/0009-0002-4942-7550"},"institutions":[{"id":"https://openalex.org/I100135526","display_name":"BGI Group (China)","ror":"https://ror.org/045pn2j94","country_code":"CN","type":"company","lineage":["https://openalex.org/I100135526"]},{"id":"https://openalex.org/I205237279","display_name":"Nankai University","ror":"https://ror.org/01y1kjr75","country_code":"CN","type":"education","lineage":["https://openalex.org/I205237279"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fei Yang","raw_affiliation_strings":["College of Computer Science, Nankai University, Tianjin, China","NKIARI, Shenzhen, Guangdong, China"],"affiliations":[{"raw_affiliation_string":"College of Computer Science, Nankai University, Tianjin, China","institution_ids":["https://openalex.org/I205237279"]},{"raw_affiliation_string":"NKIARI, Shenzhen, Guangdong, China","institution_ids":["https://openalex.org/I100135526"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039582068","display_name":"Claudio Cusano","orcid":"https://orcid.org/0000-0001-9365-8167"},"institutions":[{"id":"https://openalex.org/I25217355","display_name":"University of Pavia","ror":"https://ror.org/00s6t1f81","country_code":"IT","type":"education","lineage":["https://openalex.org/I25217355"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Claudio Cusano","raw_affiliation_strings":["Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy"],"affiliations":[{"raw_affiliation_string":"Department of Electrical, Computer and Biomedical Engineering, University of Pavia, Pavia, Italy","institution_ids":["https://openalex.org/I25217355"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5064029620","display_name":"Andrew D. Bagdanov","orcid":"https://orcid.org/0000-0001-6408-7043"},"institutions":[{"id":"https://openalex.org/I45084792","display_name":"University of Florence","ror":"https://ror.org/04jr1s763","country_code":"IT","type":"education","lineage":["https://openalex.org/I45084792"]}],"countries":["IT"],"is_corresponding":false,"raw_author_name":"Andrew D. Bagdanov","raw_affiliation_strings":["University of Florence, Media Integration and Communication Center (MICC), 50139, Firenze, Italy"],"affiliations":[{"raw_affiliation_string":"University of Florence, Media Integration and Communication Center (MICC), 50139, Firenze, Italy","institution_ids":["https://openalex.org/I45084792"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101958996","display_name":"Joost van de Weijer","orcid":"https://orcid.org/0000-0002-9656-9706"},"institutions":[{"id":"https://openalex.org/I4210122226","display_name":"Centre de Recerca Matem\u00e0tica","ror":"https://ror.org/020s51w82","country_code":"ES","type":"facility","lineage":["https://openalex.org/I4210122226","https://openalex.org/I4387153040"]},{"id":"https://openalex.org/I123044942","display_name":"Universitat Aut\u00f2noma de Barcelona","ror":"https://ror.org/052g8jq94","country_code":"ES","type":"education","lineage":["https://openalex.org/I123044942"]}],"countries":["ES"],"is_corresponding":false,"raw_author_name":"Joost van de Weijer","raw_affiliation_strings":["Computer Vision Center, 08193, Bellaterra, Cerdanyola, Barcelona, Spain","Universitat Aut\u00f2noma de Barcelona, 08193, Bellaterra, Barcelona, Spain"],"affiliations":[{"raw_affiliation_string":"Computer Vision Center, 08193, Bellaterra, Cerdanyola, Barcelona, Spain","institution_ids":["https://openalex.org/I4210122226"]},{"raw_affiliation_string":"Universitat Aut\u00f2noma de Barcelona, 08193, Bellaterra, Barcelona, Spain","institution_ids":["https://openalex.org/I123044942"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":5,"corresponding_author_ids":["https://openalex.org/A5030534371"],"corresponding_institution_ids":["https://openalex.org/I25217355"],"apc_list":{"value":2890,"currency":"EUR","value_usd":3690},"apc_paid":{"value":2890,"currency":"EUR","value_usd":3690},"fwci":2.671,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.89218856,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":99},"biblio":{"volume":"133","issue":"7","first_page":"4571","last_page":"4589"},"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":1.0,"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":1.0,"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/T12676","display_name":"Machine Learning and ELM","score":0.9983999729156494,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.9952999949455261,"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/artificial-intelligence","display_name":"Artificial intelligence","score":0.7302039861679077},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.660078227519989},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.5981476306915283},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.4981722831726074},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.48951709270477295},{"id":"https://openalex.org/keywords/class","display_name":"Class (philosophy)","score":0.4760030210018158},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.4547812044620514},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3386484682559967},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.16606634855270386},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.12202852964401245}],"concepts":[{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.7302039861679077},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.660078227519989},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5981476306915283},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4981722831726074},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.48951709270477295},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.4760030210018158},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.4547812044620514},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3386484682559967},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.16606634855270386},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.12202852964401245},{"id":"https://openalex.org/C41895202","wikidata":"https://www.wikidata.org/wiki/Q8162","display_name":"Linguistics","level":1,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","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}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1007/s11263-025-02374-x","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11263-025-02374-x","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11263-025-02374-x.pdf","source":{"id":"https://openalex.org/S25538012","display_name":"International Journal of Computer Vision","issn_l":"0920-5691","issn":["0920-5691","1573-1405"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Journal of Computer Vision","raw_type":"journal-article"},{"id":"pmh:oai:flore.unifi.it:2158/1417033","is_oa":true,"landing_page_url":"https://link.springer.com/article/10.1007/s11263-025-02374-x#citeas","pdf_url":null,"source":{"id":"https://openalex.org/S4306402033","display_name":"Florence Research (University of Florence)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I45084792","host_organization_name":"University of Florence","host_organization_lineage":["https://openalex.org/I45084792"],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"doi:10.1007/s11263-025-02374-x","is_oa":true,"landing_page_url":"https://doi.org/10.1007/s11263-025-02374-x","pdf_url":"https://link.springer.com/content/pdf/10.1007/s11263-025-02374-x.pdf","source":{"id":"https://openalex.org/S25538012","display_name":"International Journal of Computer Vision","issn_l":"0920-5691","issn":["0920-5691","1573-1405"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319900","host_organization_name":"Springer Science+Business Media","host_organization_lineage":["https://openalex.org/P4310319900","https://openalex.org/P4310319965"],"host_organization_lineage_names":["Springer Science+Business Media","Springer Nature"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"International Journal of Computer Vision","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5410076643","display_name":null,"funder_award_id":"951911","funder_id":"https://openalex.org/F4320335254","funder_display_name":"Horizon 2020"}],"funders":[{"id":"https://openalex.org/F4320315803","display_name":"Centres de Recerca de Catalunya","ror":null},{"id":"https://openalex.org/F4320322930","display_name":"Ministerio de Ciencia e Innovaci\u00f3n","ror":"https://ror.org/034900433"},{"id":"https://openalex.org/F4320335254","display_name":"Horizon 2020","ror":null}],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4408387851.pdf"},"referenced_works_count":67,"referenced_works":["https://openalex.org/W1981276685","https://openalex.org/W2037979274","https://openalex.org/W2117539524","https://openalex.org/W2194775991","https://openalex.org/W2473930607","https://openalex.org/W2551176409","https://openalex.org/W2560647685","https://openalex.org/W2737492962","https://openalex.org/W2782276075","https://openalex.org/W2786446225","https://openalex.org/W2791091755","https://openalex.org/W2804613085","https://openalex.org/W2884282566","https://openalex.org/W2898290017","https://openalex.org/W2948734064","https://openalex.org/W2952571470","https://openalex.org/W2954929116","https://openalex.org/W2963072899","https://openalex.org/W2963588172","https://openalex.org/W2963788399","https://openalex.org/W2964189064","https://openalex.org/W2972313371","https://openalex.org/W2990138404","https://openalex.org/W3000677656","https://openalex.org/W3013325675","https://openalex.org/W3017160304","https://openalex.org/W3030364939","https://openalex.org/W3034736335","https://openalex.org/W3034867292","https://openalex.org/W3035501943","https://openalex.org/W3041184466","https://openalex.org/W3091958244","https://openalex.org/W3093315565","https://openalex.org/W3094502228","https://openalex.org/W3098511564","https://openalex.org/W3107810305","https://openalex.org/W3118608800","https://openalex.org/W3138516171","https://openalex.org/W3159481202","https://openalex.org/W3163939464","https://openalex.org/W3168149265","https://openalex.org/W3170841864","https://openalex.org/W3174083140","https://openalex.org/W3177352298","https://openalex.org/W3178686235","https://openalex.org/W3180392831","https://openalex.org/W3181161034","https://openalex.org/W3203805317","https://openalex.org/W4214634256","https://openalex.org/W4214893857","https://openalex.org/W4225484930","https://openalex.org/W4225777901","https://openalex.org/W4281773771","https://openalex.org/W4292793894","https://openalex.org/W4292828582","https://openalex.org/W4312238419","https://openalex.org/W4312351187","https://openalex.org/W4312615142","https://openalex.org/W4312640688","https://openalex.org/W4387596160","https://openalex.org/W4390872344","https://openalex.org/W4402092710","https://openalex.org/W6737778391","https://openalex.org/W6739901393","https://openalex.org/W6741087337","https://openalex.org/W6757384668","https://openalex.org/W6767117800"],"related_works":["https://openalex.org/W3147584709","https://openalex.org/W2961085424","https://openalex.org/W2977677679","https://openalex.org/W1992327129","https://openalex.org/W2381986121","https://openalex.org/W2370918718","https://openalex.org/W4224009465","https://openalex.org/W2256933480","https://openalex.org/W4306674287","https://openalex.org/W2027854990"],"abstract_inverted_index":{"Abstract":[0],"Vision":[1],"transformers":[2],"(ViTs)":[3],"have":[4],"achieved":[5,70],"remarkable":[6],"successes":[7],"across":[8],"a":[9,17,36,178],"broad":[10],"range":[11],"of":[12,44,49,56,63,98,116,143,181,198,215],"computer":[13],"vision":[14],"applications.":[15],"As":[16],"consequence,":[18],"there":[19],"has":[20],"been":[21],"increasing":[22],"interest":[23],"in":[24,129,189],"extending":[25],"continual":[26,51,117],"learning":[27,52,88,193],"theory":[28],"and":[29,148,201,224],"techniques":[30],"to":[31,81,125,140,236],"ViT":[32,118,132,239],"architectures.":[33],"We":[34],"propose":[35,122,177],"new":[37,89,179,194,210],"method":[38,180,230],"for":[39,105,154,207],"exemplar-free":[40,50,229],"class":[41],"incremental":[42],"training":[43],"ViTs.":[45],"The":[46,196],"main":[47],"challenge":[48],"is":[53,68,138],"maintaining":[54],"plasticity":[55,208],"the":[57,82,96,114,127,130,144,151,163,190],"learner":[58],"without":[59],"causing":[60],"catastrophic":[61],"forgetting":[62,214],"previously":[64],"learned":[65],"tasks.":[66,90,156,195],"This":[67,135],"often":[69],"via":[71],"exemplar":[72],"replay":[73],"which":[74,85,104,167],"can":[75],"help":[76],"recalibrate":[77],"previous":[78,102,155,216],"task":[79],"classifiers":[80],"feature":[83,182,187,203],"drift":[84,128,183,188,204],"occurs":[86],"when":[87,192,234],"Exemplar":[91],"replay,":[92],"however,":[93],"comes":[94],"at":[95],"cost":[97],"retaining":[99],"samples":[100],"from":[101,170],"tasks":[103,211],"many":[106],"applications":[107],"may":[108],"not":[109,161],"be":[110],"possible.":[111],"To":[112],"address":[113],"problem":[115],"training,":[119],"we":[120,176],"first":[121],"gated":[123,158,199],"class-attention":[124,141,159,200],"minimize":[126],"final":[131],"transformer":[133,146],"block.":[134],"mask-based":[136],"gating":[137],"applied":[139],"mechanism":[142],"last":[145],"block":[147],"strongly":[149],"regulates":[150],"weights":[152],"crucial":[153],"Importantly,":[157],"does":[160],"require":[162],"task-ID":[164],"during":[165],"inference,":[166],"distinguishes":[168],"it":[169],"other":[171],"parameter":[172],"isolation":[173],"methods.":[174],"Secondly,":[175],"compensation":[184,205],"that":[185,227],"accommodates":[186],"backbone":[191],"combination":[197],"cascaded":[202],"allows":[206],"towards":[209],"while":[212],"limiting":[213],"ones.":[217],"Extensive":[218],"experiments":[219],"performed":[220],"on":[221],"CIFAR-100,":[222],"Tiny-ImageNet":[223],"ImageNet100":[225],"demonstrate":[226],"our":[228],"obtains":[231],"competitive":[232],"results":[233],"compared":[235],"rehearsal":[237],"based":[238],"methods.(Code:":[240],"https://github.com/OcraM17/GCAB-CFDC":[241],")":[242]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-03-17T17:19:04.345684","created_date":"2025-10-10T00:00:00"}