{"id":"https://openalex.org/W7133539762","doi":"https://doi.org/10.48550/arxiv.2603.02554","title":"Generalizable Knowledge Distillation from Vision Foundation Models for Semantic Segmentation","display_name":"Generalizable Knowledge Distillation from Vision Foundation Models for Semantic Segmentation","publication_year":2026,"publication_date":"2026-03-03","ids":{"openalex":"https://openalex.org/W7133539762","doi":"https://doi.org/10.48550/arxiv.2603.02554"},"language":null,"primary_location":{"id":"pmh:oai:zenodo.org:20199823","is_oa":true,"landing_page_url":"https://arxiv.org/abs/arXiv:2603.02554","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"info:eu-repo/semantics/conferencePaper"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/abs/arXiv:2603.02554","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5042003236","display_name":"Chonghua Lv","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lv, Chonghua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128125004","display_name":"Dong Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Dong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128114004","display_name":"Shuang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Shuang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125713679","display_name":"Dou Quan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Quan, Dou","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066376249","display_name":"Ning Huyan","orcid":"https://orcid.org/0000-0002-6123-8659"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huyan, Ning","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128045823","display_name":"Nicu Sebe","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sebe, Nicu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128065085","display_name":"Zhun Zhong","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhong, Zhun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":0,"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":null,"last_page":null},"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":0.7208999991416931,"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":0.7208999991416931,"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":0.16200000047683716,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.05550000071525574,"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/overfitting","display_name":"Overfitting","score":0.6814000010490417},{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.6132000088691711},{"id":"https://openalex.org/keywords/segmentation","display_name":"Segmentation","score":0.6118000149726868},{"id":"https://openalex.org/keywords/robustness","display_name":"Robustness (evolution)","score":0.5881999731063843},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5806999802589417},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5092999935150146},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.5065000057220459},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.4627000093460083},{"id":"https://openalex.org/keywords/domain-knowledge","display_name":"Domain knowledge","score":0.35429999232292175}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.762499988079071},{"id":"https://openalex.org/C22019652","wikidata":"https://www.wikidata.org/wiki/Q331309","display_name":"Overfitting","level":3,"score":0.6814000010490417},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6434000134468079},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.6132000088691711},{"id":"https://openalex.org/C89600930","wikidata":"https://www.wikidata.org/wiki/Q1423946","display_name":"Segmentation","level":2,"score":0.6118000149726868},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5983999967575073},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.5881999731063843},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5806999802589417},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5092999935150146},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.5065000057220459},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.4627000093460083},{"id":"https://openalex.org/C207685749","wikidata":"https://www.wikidata.org/wiki/Q2088941","display_name":"Domain knowledge","level":2,"score":0.35429999232292175},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.3481000065803528},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.34279999136924744},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.33809998631477356},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.33160001039505005},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3109999895095825},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3077000081539154},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.30329999327659607},{"id":"https://openalex.org/C161301231","wikidata":"https://www.wikidata.org/wiki/Q3478658","display_name":"Knowledge representation and reasoning","level":2,"score":0.2939000129699707},{"id":"https://openalex.org/C81363708","wikidata":"https://www.wikidata.org/wiki/Q17084460","display_name":"Convolutional neural network","level":2,"score":0.2809000015258789},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.271699994802475},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2606000006198883},{"id":"https://openalex.org/C28006648","wikidata":"https://www.wikidata.org/wiki/Q6934509","display_name":"Multi-task learning","level":3,"score":0.25850000977516174},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.2563999891281128}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:zenodo.org:20199823","is_oa":true,"landing_page_url":"https://arxiv.org/abs/arXiv:2603.02554","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"info:eu-repo/semantics/conferencePaper"},{"id":"doi:10.48550/arxiv.2603.02554","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.02554","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:zenodo.org:20199823","is_oa":true,"landing_page_url":"https://arxiv.org/abs/arXiv:2603.02554","pdf_url":null,"source":{"id":"https://openalex.org/S4306400562","display_name":"Zenodo (CERN European Organization for Nuclear Research)","issn_l":null,"issn":null,"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I67311998","host_organization_name":"European Organization for Nuclear Research","host_organization_lineage":["https://openalex.org/I67311998"],"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":"IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","raw_type":"info:eu-repo/semantics/conferencePaper"},"sustainable_development_goals":[{"score":0.7129276394844055,"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Knowledge":[0,64],"distillation":[1,121],"(KD)":[2],"has":[3],"been":[4],"widely":[5],"applied":[6],"in":[7,95,160,165],"semantic":[8],"segmentation":[9],"to":[10,109,129,132],"compress":[11],"large":[12],"models,":[13],"but":[14],"conventional":[15,55],"approaches":[16],"primarily":[17],"preserve":[18],"in-domain":[19],"accuracy":[20],"while":[21],"neglecting":[22],"out-of-domain":[23],"generalization,":[24],"which":[25],"is":[26],"essential":[27],"under":[28],"distribution":[29],"shifts.":[30],"This":[31],"limitation":[32],"becomes":[33],"more":[34],"severe":[35],"with":[36,54],"the":[37,82,85,96],"emergence":[38],"of":[39,158],"vision":[40],"foundation":[41],"models":[42],"(VFMs):":[43],"although":[44],"VFMs":[45],"exhibit":[46],"strong":[47],"robustness":[48],"on":[49,142],"unseen":[50],"data,":[51],"distilling":[52],"them":[53],"KD":[56,153],"often":[57],"compromises":[58],"this":[59],"ability.":[60],"We":[61],"propose":[62],"Generalizable":[63],"Distillation":[65],"(GKD),":[66],"a":[67,118],"multi-stage":[68],"framework":[69],"that":[70,148],"explicitly":[71],"enhances":[72],"generalization.":[73],"GKD":[74,149],"decouples":[75],"representation":[76],"learning":[77],"from":[78,138],"task":[79,104],"learning.":[80],"In":[81],"first":[83],"stage,":[84,98],"student":[86,124],"acquires":[87],"domain-agnostic":[88],"representations":[89,100,131],"through":[90],"selective":[91],"feature":[92],"distillation,":[93],"and":[94,163],"second":[97],"these":[99],"are":[101],"frozen":[102],"for":[103],"adaptation,":[105],"thereby":[106],"mitigating":[107],"overfitting":[108],"visible":[110],"domains.":[111],"To":[112],"further":[113],"support":[114],"transfer,":[115],"we":[116],"introduce":[117],"query-based":[119],"soft":[120],"mechanism,":[122],"where":[123],"features":[125],"act":[126],"as":[127],"queries":[128],"teacher":[130],"selectively":[133],"retrieve":[134],"transferable":[135],"spatial":[136],"knowledge":[137],"VFMs.":[139],"Extensive":[140],"experiments":[141],"five":[143],"domain":[144],"generalization":[145],"benchmarks":[146],"demonstrate":[147],"consistently":[150],"outperforms":[151],"existing":[152],"methods,":[154],"achieving":[155],"average":[156],"gains":[157],"+1.9%":[159],"foundation-to-foundation":[161],"(F2F)":[162],"+10.6%":[164],"foundation-to-local":[166],"(F2L)":[167],"distillation.":[168],"The":[169],"code":[170],"will":[171],"be":[172],"available":[173],"at":[174],"https://github.com/Younger-hua/GKD.":[175]},"counts_by_year":[],"updated_date":"2026-07-15T18:14:33.161393","created_date":"2026-03-05T00:00:00"}