{"id":"https://openalex.org/W7169794980","doi":"https://doi.org/10.48550/arxiv.2607.15485","title":"Diffusion models recover accurate mixture weights despite score function insensitivity","display_name":"Diffusion models recover accurate mixture weights despite score function insensitivity","publication_year":2026,"publication_date":"2026-07-16","ids":{"openalex":"https://openalex.org/W7169794980","doi":"https://doi.org/10.48550/arxiv.2607.15485"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2607.15485","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.15485","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.2607.15485","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5084684009","display_name":"Andrew Dennehy","orcid":"https://orcid.org/0009-0001-6893-1990"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dennehy, Andrew","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5141200908","display_name":"Ramchandran Muthukumar","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Muthukumar, Ramchandran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5141179183","display_name":"Rebecca Willett","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Willett, Rebecca","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5141176317","display_name":"Nisha Chandramoorthy","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chandramoorthy, Nisha","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":0.7202000021934509,"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":0.7202000021934509,"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/T11206","display_name":"Model Reduction and Neural Networks","score":0.030400000512599945,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10581","display_name":"Neural dynamics and brain function","score":0.020099999383091927,"subfield":{"id":"https://openalex.org/subfields/2805","display_name":"Cognitive Neuroscience"},"field":{"id":"https://openalex.org/fields/28","display_name":"Neuroscience"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/mixture-model","display_name":"Mixture model","score":0.7311000227928162},{"id":"https://openalex.org/keywords/sensitivity","display_name":"Sensitivity (control systems)","score":0.6873000264167786},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.6244999766349792},{"id":"https://openalex.org/keywords/diffusion","display_name":"Diffusion","score":0.4803999960422516},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.45910000801086426},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.45809999108314514},{"id":"https://openalex.org/keywords/function","display_name":"Function (biology)","score":0.45509999990463257},{"id":"https://openalex.org/keywords/matching","display_name":"Matching (statistics)","score":0.4253999888896942},{"id":"https://openalex.org/keywords/mode","display_name":"Mode (computer interface)","score":0.4002000093460083}],"concepts":[{"id":"https://openalex.org/C61224824","wikidata":"https://www.wikidata.org/wiki/Q2260434","display_name":"Mixture model","level":2,"score":0.7311000227928162},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.6873000264167786},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.6244999766349792},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5819000005722046},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.4803999960422516},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.45910000801086426},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.45809999108314514},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.45509999990463257},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.4332999885082245},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.4253999888896942},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.41920000314712524},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.4002000093460083},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3621000051498413},{"id":"https://openalex.org/C56672385","wikidata":"https://www.wikidata.org/wiki/Q17157111","display_name":"Mixture distribution","level":3,"score":0.3479999899864197},{"id":"https://openalex.org/C126691448","wikidata":"https://www.wikidata.org/wiki/Q2028919","display_name":"Magnitude (astronomy)","level":2,"score":0.34209999442100525},{"id":"https://openalex.org/C65660741","wikidata":"https://www.wikidata.org/wiki/Q3952743","display_name":"Score","level":2,"score":0.337799996137619},{"id":"https://openalex.org/C110121322","wikidata":"https://www.wikidata.org/wiki/Q865811","display_name":"Distribution (mathematics)","level":2,"score":0.3199000060558319},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.313400000333786},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3133000135421753},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.3084999918937683},{"id":"https://openalex.org/C167928553","wikidata":"https://www.wikidata.org/wiki/Q1376021","display_name":"Estimation theory","level":2,"score":0.3084000051021576},{"id":"https://openalex.org/C4199805","wikidata":"https://www.wikidata.org/wiki/Q2725903","display_name":"Gaussian noise","level":2,"score":0.30559998750686646},{"id":"https://openalex.org/C2779736610","wikidata":"https://www.wikidata.org/wiki/Q6884140","display_name":"Mixture theory","level":3,"score":0.29660001397132874},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.28529998660087585},{"id":"https://openalex.org/C2780428219","wikidata":"https://www.wikidata.org/wiki/Q16952335","display_name":"Cover (algebra)","level":2,"score":0.28439998626708984},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.2734000086784363},{"id":"https://openalex.org/C197055811","wikidata":"https://www.wikidata.org/wiki/Q207522","display_name":"Probability density function","level":2,"score":0.2721000015735626},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.26510000228881836},{"id":"https://openalex.org/C68710425","wikidata":"https://www.wikidata.org/wiki/Q5275442","display_name":"Diffusion process","level":3,"score":0.2621000111103058},{"id":"https://openalex.org/C186603090","wikidata":"https://www.wikidata.org/wiki/Q55505926","display_name":"Distribution function","level":2,"score":0.2612000107765198},{"id":"https://openalex.org/C63002673","wikidata":"https://www.wikidata.org/wiki/Q2260590","display_name":"Scoring rule","level":2,"score":0.25529998540878296}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2607.15485","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.15485","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.2607.15485","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2607.15485","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Score-based":[0],"generative":[1],"models":[2],"exhibit":[3],"a":[4,15,112,190],"puzzling":[5],"behavior:":[6],"they":[7],"often":[8],"appear":[9],"to":[10,23,50,70,109,210],"cover":[11],"all":[12],"modes":[13],"of":[14,127,168,174,202,226,230],"target":[16,66,129,232],"multimodal":[17],"distribution":[18,130],"and":[19,182],"yet":[20],"may":[21],"fail":[22],"learn":[24],"the":[25,44,51,65,77,81,90,95,102,105,118,121,125,128,146,153,157,166,171,200,219,224,231],"correct":[26],"relative":[27,108],"mode":[28,211],"amplitudes,":[29],"which":[30,124],"can":[31,75,131,205],"be":[32,132],"interpreted":[33],"as":[34,101,156],"mixture":[35,55,71,147,194,217],"weights.":[36,91,195],"We":[37,60,114],"resolve":[38],"this":[39],"apparent":[40],"paradox":[41],"by":[42],"relating":[43],"diffusion":[45,96,207],"score":[46,67,97],"matching":[47],"(DSM)":[48],"loss":[49,107,159],"error":[52],"in":[53,104,111,140],"estimating":[54],"weights":[56,78],"from":[57,134],"generated":[58,73,135],"samples.":[59,136],"show":[61,116,165,198],"that,":[62],"even":[63],"when":[64],"is":[68],"insensitive":[69],"weights,":[72,218],"samples":[74],"recover":[76],"accurately":[79],"if":[80],"scores":[82],"at":[83],"intermediate":[84],"noise":[85,180,203],"levels":[86],"are":[87,151],"informative":[88],"about":[89],"Accordingly,":[92],"we":[93,143,164,197,214],"define":[94],"sensitivity":[98,169,185,221],"index":[99],"(DSSI)":[100],"variation":[103],"DSM":[106,158],"changes":[110],"parameter.":[113],"then":[115],"that":[117,145,183,199],"DSSI":[119],"governs":[120,223],"accuracy":[122],"with":[123],"parameter":[126,229],"estimated":[133],"For":[137],"Gaussian":[138],"mixtures":[139],"arbitrary":[141],"dimensions,":[142],"prove":[144],"weight":[148],"estimation":[149],"errors":[150],"on":[152,216],"same":[154],"order":[155],"under":[160,178],"mild":[161],"conditions.":[162],"Empirically,":[163],"emergence":[167],"during":[170],"noising":[172],"process":[173],"benchmark":[175],"data":[176],"distributions":[177],"typical":[179],"schedules,":[181],"these":[184],"values":[186],"predict":[187],"how":[188],"well":[189],"well-trained":[191],"model":[192],"recovers":[193],"Furthermore,":[196],"choice":[201],"schedule":[204],"reduce":[206],"sensitivity,":[208],"leading":[209],"amplification.":[212],"Although":[213],"focus":[215],"proposed":[220],"framework":[222],"recovery":[225],"any":[227],"qualitative":[228],"distribution.":[233]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-07-21T00:00:00"}