{"id":"https://openalex.org/W7160050199","doi":"https://doi.org/10.1109/iccv51701.2025.01064","title":"Unraveling the Smoothness Properties of Diffusion Models: A Gaussian Mixture Perspective","display_name":"Unraveling the Smoothness Properties of Diffusion Models: A Gaussian Mixture Perspective","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W7160050199","doi":"https://doi.org/10.1109/iccv51701.2025.01064"},"language":"en","primary_location":{"id":"doi:10.1109/iccv51701.2025.01064","doi":"https://doi.org/10.1109/iccv51701.2025.01064","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.01064","pdf_url":null,"source":{"id":"https://openalex.org/S4363607764","display_name":"IEEE/CVF International Conference on Computer Vision (ICCV)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},"volume":null,"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2405.16418","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5035905957","display_name":"Yingyu Liang","orcid":"https://orcid.org/0000-0002-3644-774X","observed_orcids":["https://orcid.org/0000-0002-3644-774X"]},"institutions":[{"id":"https://openalex.org/I889458895","display_name":"University of Hong Kong","ror":"https://ror.org/02zhqgq86","country_code":"HK","type":"education","lineage":["https://openalex.org/I889458895"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Yingyu Liang","raw_affiliation_strings":["The University of Hong Kong"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The University of Hong Kong","institution_ids":["https://openalex.org/I889458895"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5110738653","display_name":"Zhizhou Sha","orcid":"https://orcid.org/0009-0004-2494-1945","observed_orcids":["https://orcid.org/0009-0004-2494-1945"]},"institutions":[{"id":"https://openalex.org/I86519309","display_name":"The University of Texas at Austin","ror":"https://ror.org/00hj54h04","country_code":"US","type":"education","lineage":["https://openalex.org/I86519309"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhizhou Sha","raw_affiliation_strings":["University of Texas at Austin"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Texas at Austin","institution_ids":["https://openalex.org/I86519309"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5069272192","display_name":"Zhenmei Shi","orcid":"https://orcid.org/0009-0007-6741-7598","observed_orcids":["https://orcid.org/0009-0007-6741-7598"]},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhenmei Shi","raw_affiliation_strings":["University of Wisconsin-Madison"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Wisconsin-Madison","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135235127","display_name":"Zhao Song","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I95457486","display_name":"University of California, Berkeley","ror":"https://ror.org/01an7q238","country_code":"US","type":"education","lineage":["https://openalex.org/I95457486"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhao Song","raw_affiliation_strings":["University of California,Berkeley"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of California,Berkeley","institution_ids":["https://openalex.org/I95457486"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111638387","display_name":"Mingda Wan","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mingda Wan","raw_affiliation_strings":["Independent Researcher"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Independent Researcher","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101418125","display_name":"Yufa Zhou","orcid":"https://orcid.org/0000-0001-9027-1792","observed_orcids":["https://orcid.org/0000-0001-9027-1792"]},"institutions":[{"id":"https://openalex.org/I79576946","display_name":"University of Pennsylvania","ror":"https://ror.org/00b30xv10","country_code":"US","type":"education","lineage":["https://openalex.org/I79576946"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yufa Zhou","raw_affiliation_strings":["University of Pennsylvania"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Pennsylvania","institution_ids":["https://openalex.org/I79576946"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":5,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.61391226,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":89,"max":94},"biblio":{"volume":null,"issue":null,"first_page":"11436","last_page":"11446"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.23100000619888306,"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/T12814","display_name":"Gaussian Processes and Bayesian Inference","score":0.23100000619888306,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.21559999883174896,"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/T10688","display_name":"Image and Signal Denoising Methods","score":0.11320000141859055,"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/diffusion-models","display_name":"diffusion models","score":0.9929999709129333},{"id":"https://openalex.org/keywords/gaussian-mixture-model","display_name":"Gaussian mixture model","score":0.8159999847412109}],"concepts":[{"id":"https://openalex.org/C102634674","wikidata":"https://www.wikidata.org/wiki/Q868473","display_name":"Smoothness","level":2,"score":0.5842999815940857},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.517300009727478},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.49300000071525574},{"id":"https://openalex.org/C121864883","wikidata":"https://www.wikidata.org/wiki/Q677916","display_name":"Statistical physics","level":1,"score":0.4885999858379364},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.4855000078678131},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.43290001153945923},{"id":"https://openalex.org/C134306372","wikidata":"https://www.wikidata.org/wiki/Q7754","display_name":"Mathematical analysis","level":1,"score":0.35580000281333923},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.3549000024795532},{"id":"https://openalex.org/C61224824","wikidata":"https://www.wikidata.org/wiki/Q2260434","display_name":"Mixture model","level":2,"score":0.31859999895095825},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.3183000087738037},{"id":"https://openalex.org/C61326573","wikidata":"https://www.wikidata.org/wiki/Q1496376","display_name":"Gaussian process","level":3,"score":0.31310001015663147},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.2563000023365021}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/iccv51701.2025.01064","doi":"https://doi.org/10.1109/iccv51701.2025.01064","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.01064","pdf_url":null,"source":{"id":"https://openalex.org/S4363607764","display_name":"IEEE/CVF International Conference on Computer Vision (ICCV)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2405.16418","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2405.16418","pdf_url":"https://arxiv.org/pdf/2405.16418","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2405.16418","doi":"https://doi.org/10.48550/arxiv.2405.16418","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2405.16418","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":"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2405.16418","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2405.16418","pdf_url":"https://arxiv.org/pdf/2405.16418","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[{"id":"https://openalex.org/G1567685964","display_name":"CAREER: Towards Better Understanding, Robustness, and Efficiency of Deep Learning","funder_award_id":"2046710","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G5632738584","display_name":"TRIPODS: Institute for Foundations of Data Science","funder_award_id":"2023239","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7160050199.pdf","grobid_xml":"https://content.openalex.org/works/W7160050199.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W1763164889","https://openalex.org/W1901129140","https://openalex.org/W1956647075","https://openalex.org/W1974088667","https://openalex.org/W1985792472","https://openalex.org/W1991111872","https://openalex.org/W2026302946","https://openalex.org/W2033178790","https://openalex.org/W2137854958","https://openalex.org/W2942689850","https://openalex.org/W3110212018","https://openalex.org/W3110798296","https://openalex.org/W3155072588","https://openalex.org/W4213251304","https://openalex.org/W4302366875","https://openalex.org/W4312933868","https://openalex.org/W4390872297","https://openalex.org/W4402916224","https://openalex.org/W4403841676","https://openalex.org/W4415798361","https://openalex.org/W7133215338","https://openalex.org/W7133219678","https://openalex.org/W7133224911","https://openalex.org/W7133227128"],"related_works":["https://openalex.org/W4381573345","https://openalex.org/W4396821238","https://openalex.org/W4403580047","https://openalex.org/W4416119075","https://openalex.org/W4392488191","https://openalex.org/W4416707190","https://openalex.org/W4317539811","https://openalex.org/W4323362754","https://openalex.org/W4387076256","https://openalex.org/W4403799699"],"abstract_inverted_index":{"Diffusion":[0],"models":[1],"have":[2],"made":[3],"rapid":[4],"progress":[5],"in":[6,142],"generating":[7],"high-quality":[8],"samples":[9],"across":[10],"various":[11,129],"domains.":[12],"However,":[13],"a":[14,41,57,64,82,96],"theoretical":[15,162],"understanding":[16],"of":[17,25,44,59,84,88,98,116,119,144,167],"the":[18,26,49,52,78,86,89,107,117,145,153,165,168],"Lipschitz":[19,108],"continuity":[20],"and":[21,110,134,149,155],"second":[22,111],"momentum":[23,112],"properties":[24,47],"diffusion":[27,91,130,169],"process":[28,92,170],"is":[29,56,81],"still":[30],"lacking.":[31],"In":[32],"this":[33,37],"paper,":[34],"we":[35,124],"bridge":[36],"gap":[38],"by":[39],"providing":[40],"detailed":[42],"examination":[43],"these":[45],"smoothness":[46],"for":[48,67],"case":[50],"where":[51],"target":[53,79,154],"data":[54,173],"distribution":[55,80],"mixture":[58,120],"Gaussians,":[60,85],"which":[61],"serves":[62],"as":[63,71],"universal":[65],"approximator":[66],"smooth":[68],"densities":[69],"such":[70],"image":[72],"data.":[73],"We":[74,100],"prove":[75],"that":[76,113],"if":[77],"$k$-mixture":[83,97],"density":[87],"entire":[90],"will":[93],"also":[94],"be":[95],"Gaussians.":[99],"then":[101],"derive":[102],"tight":[103],"upper":[104],"bounds":[105],"on":[106],"constant":[109],"are":[114],"independent":[115],"number":[118],"components":[121],"$k$.":[122],"Finally,":[123],"apply":[125],"our":[126],"analysis":[127],"to":[128,137],"solvers,":[131],"both":[132],"SDE":[133],"ODE":[135],"based,":[136],"establish":[138],"concrete":[139],"error":[140],"guarantees":[141],"terms":[143],"total":[146],"variation":[147],"distance":[148],"KL":[150],"divergence":[151],"between":[152],"learned":[156],"distributions.":[157,174],"Our":[158],"results":[159],"provide":[160],"deeper":[161],"insights":[163],"into":[164],"dynamics":[166],"under":[171],"common":[172]},"counts_by_year":[{"year":2024,"cited_by_count":1}],"updated_date":"2026-10-14T16:39:01.820549","created_date":"2026-05-05T00:00:00"}