{"id":"https://openalex.org/W7161573110","doi":"https://doi.org/10.48550/arxiv.2605.15309","title":"One Pass Is Not Enough: Recursive Latent Refinement for Generative Models","display_name":"One Pass Is Not Enough: Recursive Latent Refinement for Generative Models","publication_year":2026,"publication_date":"2026-05-14","ids":{"openalex":"https://openalex.org/W7161573110","doi":"https://doi.org/10.48550/arxiv.2605.15309"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.15309","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15309","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":"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.2605.15309","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136420857","display_name":"Mehdi Esmaeilzadeh","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Esmaeilzadeh, Mehdi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136438819","display_name":"Alexia Jolicoeur-Martineau","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jolicoeur-Martineau, Alexia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075265579","display_name":"Chirag Vashist","orcid":"https://orcid.org/0000-0003-3270-1533"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vashist, Chirag","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136435154","display_name":"Ke Li","orcid":"https://orcid.org/0000-0002-1318-3630"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Ke","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":0.9200999736785889,"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":0.9200999736785889,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.026799999177455902,"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":0.008700000122189522,"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/mode","display_name":"Mode (computer interface)","score":0.6344000101089478},{"id":"https://openalex.org/keywords/fidelity","display_name":"Fidelity","score":0.5795000195503235},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.5277000069618225},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.5105999708175659},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.4991999864578247},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.43639999628067017},{"id":"https://openalex.org/keywords/precision-and-recall","display_name":"Precision and recall","score":0.3962000012397766},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.3734000027179718},{"id":"https://openalex.org/keywords/high-fidelity","display_name":"High fidelity","score":0.3366999924182892}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7346000075340271},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.6344000101089478},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.5795000195503235},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.5277000069618225},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.5105999708175659},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.4991999864578247},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4961000084877014},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.43639999628067017},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4009999930858612},{"id":"https://openalex.org/C81669768","wikidata":"https://www.wikidata.org/wiki/Q2359161","display_name":"Precision and recall","level":2,"score":0.3962000012397766},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39309999346733093},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.3734000027179718},{"id":"https://openalex.org/C113364801","wikidata":"https://www.wikidata.org/wiki/Q26674","display_name":"High fidelity","level":2,"score":0.3366999924182892},{"id":"https://openalex.org/C100660578","wikidata":"https://www.wikidata.org/wiki/Q18733","display_name":"Recall","level":2,"score":0.32109999656677246},{"id":"https://openalex.org/C2777303404","wikidata":"https://www.wikidata.org/wiki/Q759757","display_name":"Convergence (economics)","level":2,"score":0.320499986410141},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.3109999895095825},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.29829999804496765},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.2955999970436096},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.28139999508857727},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2777000069618225},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.27239999175071716},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.26820001006126404},{"id":"https://openalex.org/C96250715","wikidata":"https://www.wikidata.org/wiki/Q965330","display_name":"Estimation","level":2,"score":0.25859999656677246},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.2533999979496002},{"id":"https://openalex.org/C2781316041","wikidata":"https://www.wikidata.org/wiki/Q1230584","display_name":"Diversity (politics)","level":2,"score":0.250900000333786}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.15309","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15309","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":"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.2605.15309","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.15309","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":"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":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Despite":[0],"remarkable":[1],"progress,":[2],"image":[3],"generation":[4],"is":[5,20,71,78,185],"far":[6],"from":[7],"solved.":[8],"The":[9],"dominant":[10],"metric,":[11],"FID,":[12,37,66,157],"conflates":[13],"sample":[14,101],"fidelity":[15],"with":[16,115,131,158],"mode":[17,31,93,139],"coverage":[18,140],"and":[19,60,67,82,120,128,148,165,173,177,207],"close":[21],"to":[22,65,79,188],"being":[23],"saturated.":[24],"Yet":[25],"a":[26,35,39,47,88],"model":[27,48,89],"can":[28,45],"still":[29],"exhibit":[30],"collapse":[32],"while":[33,154],"achieving":[34],"low":[36],"since":[38],"handful":[40],"of":[41,200],"sharp,":[42],"near-duplicate":[43],"images":[44],"outscore":[46],"that":[49,58,68,90,122,182,193],"faithfully":[50],"covers":[51],"the":[52,74,108,145,183,198],"full":[53],"data":[54],"distribution.":[55],"We":[56,103],"argue":[57],"precision":[59,147],"recall":[61,86,149],"are":[62],"essential":[63],"complements":[64],"because":[69],"FID":[70,196],"already":[72],"saturated,":[73],"more":[75],"meaningful":[76],"goal":[77],"improve":[80],"diversity":[81,208],"coverage.":[83],"Achieving":[84],"high":[85],"requires":[87],"explicitly":[91],"prioritizes":[92],"coverage,":[94,201],"unlike":[95],"most":[96],"generative":[97],"models,":[98],"which":[99,106,137],"optimize":[100],"fidelity.":[102],"introduce":[104],"RTM,":[105],"replaces":[107],"single-pass":[109],"latent":[110],"mapping":[111],"in":[112],"style-based":[113],"generators":[114],"an":[116],"iterative":[117],"refinement":[118,203],"process,":[119],"show":[121],"this":[123],"consistently":[124],"improves":[125,171,204],"both":[126,205],"quality":[127,206],"diversity.":[129],"Integrated":[130],"Implicit":[132],"Maximum":[133],"Likelihood":[134],"Estimation":[135],"(IMLE),":[136],"optimizes":[138],"by":[141],"design,":[142],"RTM":[143,169],"achieves":[144],"highest":[146],"among":[150],"current":[151],"state-of-the-art":[152],"approaches":[153],"maintaining":[155],"competitive":[156,195],"improvements":[159],"across":[160],"CIFAR-10,":[161],"CelebA-HQ":[162],"at":[163,179,197],"256x256,":[164],"nine":[166],"few-shot":[167],"benchmarks.":[168],"also":[170],"StyleGAN2":[172],"StyleGAN2-ADA":[174],"on":[175],"CIFAR-10":[176],"AFHQ-v1":[178],"512x512,":[180],"demonstrating":[181],"benefit":[184],"not":[186],"specific":[187],"IMLE.":[189],"Unlike":[190],"flow-matching":[191],"baselines":[192],"achieve":[194],"expense":[199],"recursive":[202],"simultaneously.":[209]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-19T00:00:00"}