{"id":"https://openalex.org/W4415964433","doi":"https://doi.org/10.48550/arxiv.2510.17421","title":"Diffusion Models as Dataset Distillation Priors","display_name":"Diffusion Models as Dataset Distillation Priors","publication_year":2025,"publication_date":"2025-10-20","ids":{"openalex":"https://openalex.org/W4415964433","doi":"https://doi.org/10.48550/arxiv.2510.17421"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2510.17421","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.17421","pdf_url":"https://arxiv.org/pdf/2510.17421","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2510.17421","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5087092119","display_name":"Duo Su","orcid":"https://orcid.org/0000-0002-9607-3639","observed_orcids":["https://orcid.org/0000-0002-9607-3639"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Su, Duo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Wu, Huyu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Huyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074300516","display_name":"Huanran Chen","orcid":"https://orcid.org/0000-0003-0847-9485","observed_orcids":["https://orcid.org/0000-0003-0847-9485"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Huanran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111325456","display_name":"Yiming Shi","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shi, Yiming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100651940","display_name":"Yuzhu Wang","orcid":"https://orcid.org/0000-0003-4305-4159","observed_orcids":["https://orcid.org/0000-0003-4305-4159"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yuzhu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100355602","display_name":"Xi Ye","orcid":"https://orcid.org/0000-0003-0331-7946","observed_orcids":["https://orcid.org/0000-0003-0331-7946"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ye, Xi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100370162","display_name":"Jun Zhu","orcid":"https://orcid.org/0000-0003-1938-4705","observed_orcids":["https://orcid.org/0000-0003-1938-4705"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhu, Jun","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":1,"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":0.9672999978065491,"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":0.9672999978065491,"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/T12535","display_name":"Machine Learning and Data Classification","score":0.01140000019222498,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.007000000216066837,"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/dataset-distillation","display_name":"dataset distillation","score":0.9940000176429749},{"id":"https://openalex.org/keywords/diffusion-models","display_name":"diffusion models","score":0.9629999995231628},{"id":"https://openalex.org/keywords/mercer-kernels","display_name":"Mercer kernels","score":0.9449999928474426},{"id":"https://openalex.org/keywords/cross-architecture-generalization","display_name":"cross-architecture generalization","score":0.9240000247955322},{"id":"https://openalex.org/keywords/imagenet-1k","display_name":"ImageNet-1K","score":0.7770000100135803}],"concepts":[{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.7778000235557556},{"id":"https://openalex.org/C37381756","wikidata":"https://www.wikidata.org/wiki/Q20203288","display_name":"Representativeness heuristic","level":2,"score":0.7520999908447266},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6279000043869019},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.5221999883651733},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.4925000071525574},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.4853000044822693},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.44830000400543213},{"id":"https://openalex.org/C61224824","wikidata":"https://www.wikidata.org/wiki/Q2260434","display_name":"Mixture model","level":2,"score":0.4203999936580658},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3871999979019165},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.36640000343322754},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.3549000024795532},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3472999930381775},{"id":"https://openalex.org/C160920958","wikidata":"https://www.wikidata.org/wiki/Q7662746","display_name":"Synthetic data","level":2,"score":0.334199994802475},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3312000036239624},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.3273000121116638},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.325300008058548},{"id":"https://openalex.org/C31170391","wikidata":"https://www.wikidata.org/wiki/Q188619","display_name":"Hierarchy","level":2,"score":0.2874999940395355},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.2865999937057495},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.272599995136261},{"id":"https://openalex.org/C55037315","wikidata":"https://www.wikidata.org/wiki/Q5421151","display_name":"Experimental data","level":2,"score":0.25690001249313354}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2510.17421","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.17421","pdf_url":"https://arxiv.org/pdf/2510.17421","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2510.17421","doi":"https://doi.org/10.48550/arxiv.2510.17421","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2510.17421","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,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2510.17421","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2510.17421","pdf_url":"https://arxiv.org/pdf/2510.17421","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4414581237","https://openalex.org/W7137819731","https://openalex.org/W4413277992","https://openalex.org/W7197040798","https://openalex.org/W7138133176","https://openalex.org/W7147393827","https://openalex.org/W4386075990","https://openalex.org/W7148415055","https://openalex.org/W7113072886","https://openalex.org/W6891867257"],"abstract_inverted_index":{"Dataset":[0],"distillation":[1,36,161],"aims":[2],"to":[3,65,104],"synthesize":[4],"compact":[5],"yet":[6],"informative":[7],"datasets":[8,139],"from":[9],"large":[10],"ones.":[11],"A":[12],"significant":[13],"challenge":[14],"in":[15,27,50,90,136],"this":[16,100],"field":[17],"is":[18,53],"achieving":[19,141],"a":[20,28,94,150,165],"trifecta":[21],"of":[22,62,113,159,173],"diversity,":[23],"generalization,":[24],"and":[25,87,127,156],"representativeness":[26,48,80,112],"single":[29],"distilled":[30,114,175],"dataset.":[31,176],"Although":[32],"recent":[33],"generative":[34],"dataset":[35,160],"methods":[37,135],"adopt":[38],"powerful":[39],"diffusion":[40,51,108,154],"models":[41,52],"as":[42,102,125],"their":[43],"foundation":[44],"models,":[45],"the":[46,60,83,106,111,157,171,174],"inherent":[47],"prior":[49,101],"overlooked.":[54],"Consequently,":[55],"these":[56],"approaches":[57],"often":[58],"necessitate":[59],"integration":[61],"external":[63],"constraints":[64],"enhance":[66],"data":[67,89],"quality.":[68],"To":[69],"address":[70],"this,":[71],"we":[72],"propose":[73],"Diffusion":[74],"As":[75],"Priors":[76],"(DAP),":[77],"which":[78],"formalizes":[79],"by":[81],"quantifying":[82],"similarity":[84],"between":[85,153],"synthetic":[86],"real":[88],"feature":[91],"space":[92],"using":[93],"Mercer":[95],"kernel.":[96],"We":[97],"then":[98],"introduce":[99],"guidance":[103],"steer":[105],"reverse":[107],"process,":[109],"enhancing":[110],"samples":[115],"without":[116],"any":[117],"retraining.":[118],"Extensive":[119],"experiments":[120],"on":[121],"large-scale":[122],"datasets,":[123],"such":[124],"ImageNet-1K":[126],"its":[128],"subsets,":[129],"demonstrate":[130],"that":[131],"DAP":[132],"outperforms":[133],"state-of-the-art":[134],"generating":[137],"high-fidelity":[138],"while":[140],"superior":[142],"cross-architecture":[143],"generalization.":[144],"Our":[145],"work":[146],"not":[147],"only":[148],"establishes":[149],"theoretical":[151],"connection":[152],"priors":[155],"objectives":[158],"but":[162],"also":[163],"provides":[164],"practical,":[166],"training-free":[167],"framework":[168],"for":[169],"improving":[170],"quality":[172]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2025-10-22T00:00:00"}