{"id":"https://openalex.org/W7163710957","doi":"https://doi.org/10.48550/arxiv.2606.06078","title":"Knowledge Distillation for Visual Autoregressive Models","display_name":"Knowledge Distillation for Visual Autoregressive Models","publication_year":2026,"publication_date":"2026-06-04","ids":{"openalex":"https://openalex.org/W7163710957","doi":"https://doi.org/10.48550/arxiv.2606.06078"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.06078","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06078","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.06078","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5004548591","display_name":"Elia Peruzzo","orcid":"https://orcid.org/0000-0003-0119-3783"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peruzzo, Elia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016712614","display_name":"Aritra Bhowmik","orcid":"https://orcid.org/0000-0002-0155-1408"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bhowmik, Aritra","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5010909962","display_name":"Guillaume Sauti\u00e8re","orcid":"https://orcid.org/0000-0002-8795-3335"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sautiere, Guillaume","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137986221","display_name":"Yuki M Asano","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Asano, Yuki M","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5045620960","display_name":"Amirhossein Habibian","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Habibian, Amirhossein","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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.5608000159263611,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.5608000159263611,"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":0.23010000586509705,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.025299999862909317,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/distillation","display_name":"Distillation","score":0.8458999991416931},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.8201000094413757},{"id":"https://openalex.org/keywords/ambiguity","display_name":"Ambiguity","score":0.5559999942779541},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.3889000117778778},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.33869999647140503},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.3361999988555908}],"concepts":[{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.8458999991416931},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.8201000094413757},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7034000158309937},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5848000049591064},{"id":"https://openalex.org/C2780522230","wikidata":"https://www.wikidata.org/wiki/Q1140419","display_name":"Ambiguity","level":2,"score":0.5559999942779541},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5214999914169312},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3889000117778778},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.33869999647140503},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.3361999988555908},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3203999996185303},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.31869998574256897},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.311599999666214},{"id":"https://openalex.org/C154030694","wikidata":"https://www.wikidata.org/wiki/Q1436074","display_name":"Fractionating column","level":3,"score":0.2856000065803528},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2524999976158142}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.06078","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06078","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2606.06078","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.06078","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.8260747194290161}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Autoregressive":[0],"(AR)":[1],"image":[2,56],"generation":[3,39],"models":[4,107],"are":[5],"highly":[6],"expressive":[7],"but":[8],"computationally":[9],"intensive,":[10],"motivating":[11],"effective":[12],"model":[13,23],"compression.":[14],"Knowledge":[15],"distillation":[16,52,64,102,135],"(KD)":[17],"is":[18],"a":[19,101],"natural":[20],"approach":[21],"for":[22,54,72,104],"compression":[24],"and":[25,83,118],"has":[26],"been":[27],"widely":[28],"studied":[29],"in":[30,36],"language":[31,73],"modeling,":[32],"yet":[33],"its":[34],"behavior":[35],"visual":[37,84,105],"AR":[38,55,127],"remains":[40],"underexplored.":[41],"In":[42],"this":[43],"work,":[44],"we":[45,98],"present":[46],"the":[47,138],"first":[48],"systematic":[49],"study":[50],"of":[51],"strategies":[53],"models.":[57,142],"Our":[58],"analysis":[59],"shows":[60],"that":[61,108,130],"while":[62,113],"standard":[63],"can":[65],"yield":[66],"meaningful":[67],"gains,":[68],"recent":[69],"methods":[70],"developed":[71],"do":[74],"not":[75],"directly":[76],"transfer":[77],"to":[78,140],"images:":[79],"long":[80],"decoding":[81],"horizons":[82],"token":[85],"ambiguity":[86],"make":[87],"teacher":[88,116],"supervision":[89,117],"unreliable":[90],"especially":[91],"under":[92],"student-conditioned":[93],"contexts.":[94],"To":[95],"address":[96],"this,":[97],"propose":[99],"VarKD,":[100],"framework":[103],"autoregressive":[106],"distills":[109],"on":[110,123],"student":[111],"samples":[112],"selectively":[114],"applying":[115],"reducing":[119],"token-level":[120],"ambiguity.":[121],"Experiments":[122],"ImageNet":[124],"across":[125],"multiple":[126],"backbones":[128],"show":[129],"VarKD":[131],"consistently":[132],"outperforms":[133],"prior":[134],"baselines,":[136],"narrowing":[137],"gap":[139],"large-scale":[141]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-06T00:00:00"}