{"id":"https://openalex.org/W4308163867","doi":"https://doi.org/10.48550/arxiv.2211.01324","title":"eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers","display_name":"eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers","publication_year":2022,"publication_date":"2022-11-02","ids":{"openalex":"https://openalex.org/W4308163867","doi":"https://doi.org/10.48550/arxiv.2211.01324"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2211.01324","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2211.01324","pdf_url":"https://arxiv.org/pdf/2211.01324","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2211.01324","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5068290472","display_name":"Yogesh Balaji","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Balaji, Yogesh","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5058394307","display_name":"Seungjun Nah","orcid":"https://orcid.org/0000-0003-3971-9402"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Nah, Seungjun","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5021345568","display_name":"Xun Huang","orcid":"https://orcid.org/0000-0001-9441-1473"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Xun","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038984764","display_name":"Arash Vahdat","orcid":"https://orcid.org/0009-0005-9476-1306"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Vahdat, Arash","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008049062","display_name":"Jiaming Song","orcid":"https://orcid.org/0000-0002-1931-1699"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Jiaming","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Zhang, Qinsheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Qinsheng","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5032255237","display_name":"Karsten Kreis","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kreis, Karsten","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5048526734","display_name":"Miika Aittala","orcid":"https://orcid.org/0000-0003-0988-5397"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aittala, Miika","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5080133323","display_name":"Timo Aila","orcid":"https://orcid.org/0000-0002-9437-4438"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Aila, Timo","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5029831111","display_name":"Samuli Laine","orcid":"https://orcid.org/0000-0002-0903-3197"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Laine, Samuli","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066242985","display_name":"Bryan Catanzaro","orcid":"https://orcid.org/0000-0003-0034-7728"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Catanzaro, Bryan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004093189","display_name":"Tero Karras","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Karras, Tero","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5100335524","display_name":"Ming-Yu Liu","orcid":"https://orcid.org/0000-0002-2239-8975"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Ming-Yu","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":13,"corresponding_author_ids":["https://openalex.org/A5068290472"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":false,"cited_by_count":223,"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":0.9034000039100647,"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":0.9034000039100647,"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/computer-science","display_name":"Computer science","score":0.7592931985855103},{"id":"https://openalex.org/keywords/embedding","display_name":"Embedding","score":0.5599930286407471},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.5463593006134033},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5250285267829895},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.500244140625},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.48645755648612976},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.48288726806640625},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.47961270809173584},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.4771527647972107},{"id":"https://openalex.org/keywords/scale","display_name":"Scale (ratio)","score":0.4292517900466919},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.30283766984939575}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7592931985855103},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.5599930286407471},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.5463593006134033},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5250285267829895},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.500244140625},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.48645755648612976},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48288726806640625},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.47961270809173584},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.4771527647972107},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.4292517900466919},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.30283766984939575},{"id":"https://openalex.org/C13280743","wikidata":"https://www.wikidata.org/wiki/Q131089","display_name":"Geodesy","level":1,"score":0.0},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.0},{"id":"https://openalex.org/C205649164","wikidata":"https://www.wikidata.org/wiki/Q1071","display_name":"Geography","level":0,"score":0.0},{"id":"https://openalex.org/C62520636","wikidata":"https://www.wikidata.org/wiki/Q944","display_name":"Quantum mechanics","level":1,"score":0.0},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.0}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2211.01324","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2211.01324","pdf_url":"https://arxiv.org/pdf/2211.01324","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},{"id":"doi:10.48550/arxiv.2211.01324","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2211.01324","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":"article"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2211.01324","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2211.01324","pdf_url":"https://arxiv.org/pdf/2211.01324","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":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"","raw_type":"text"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4365211920","https://openalex.org/W3014948380","https://openalex.org/W4380551139","https://openalex.org/W4317695495","https://openalex.org/W4287117424","https://openalex.org/W4387506531","https://openalex.org/W4238433571","https://openalex.org/W3174044702","https://openalex.org/W2967848559","https://openalex.org/W4299831724"],"abstract_inverted_index":{"Large-scale":[0],"diffusion-based":[1],"generative":[2],"models":[3,20,97,119,161],"have":[4],"led":[5],"to":[6,54,85,90,172,197,217,248],"breakthroughs":[7],"in":[8,24,45,83,140,238,245,261],"text-conditioned":[9],"high-resolution":[10],"image":[11,188,203,216,260],"synthesis.":[12],"Starting":[13],"from":[14],"random":[15],"noise,":[16],"such":[17],"text-to-image":[18,95,159,220],"diffusion":[19,96,135,160],"gradually":[21],"synthesize":[22],"images":[23],"an":[25,92,206],"iterative":[26,129],"fashion":[27],"while":[28,58,144],"conditioning":[29,62],"on":[30,50,162],"text":[31,52,61,142,241],"prompts.":[32],"We":[33,190],"find":[34],"that":[35,69,120,192,227],"their":[36],"synthesis":[37,101],"behavior":[38],"qualitatively":[39],"changes":[40],"throughout":[41,73],"this":[42],"process:":[43],"Early":[44],"sampling,":[46],"generation":[47,76,130],"strongly":[48],"relies":[49],"the":[51,60,74,124,128,146,163,181,201,211,218,236,239,250,258],"prompt":[53],"generate":[55],"text-aligned":[56],"content,":[57],"later,":[59],"is":[63,114,253,266],"almost":[64],"entirely":[65],"ignored.":[66],"This":[67],"suggests":[68],"sharing":[70],"model":[71,171],"parameters":[72],"entire":[75],"process":[77],"may":[78],"not":[79],"be":[80],"ideal.":[81],"Therefore,":[82],"contrast":[84],"existing":[86],"works,":[87],"we":[88,107,168,223],"propose":[89],"train":[91,109,169],"ensemble":[93,133],"of":[94,127,134,176,209,213],"specialized":[98,118],"for":[99,123,178,256],"different":[100,194,198],"stages.":[102],"To":[103],"maintain":[104],"training":[105],"efficiency,":[106],"initially":[108],"a":[110,174,214,225,246],"single":[111],"model,":[112],"which":[113,252],"then":[115],"split":[116],"into":[117],"are":[121],"trained":[122],"specific":[125],"stages":[126],"process.":[131],"Our":[132],"models,":[136],"called":[137],"eDiff-I,":[138],"results":[139],"improved":[141],"alignment":[143],"maintaining":[145],"same":[147],"inference":[148],"computation":[149],"cost":[150],"and":[151,186,242],"preserving":[152],"high":[153],"visual":[154],"quality,":[155],"outperforming":[156],"previous":[157],"large-scale":[158],"standard":[164],"benchmark.":[165],"In":[166],"addition,":[167],"our":[170],"exploit":[173],"variety":[175],"embeddings":[177,195],"conditioning,":[179],"including":[180],"T5":[182],"text,":[183,185],"CLIP":[184,187,202],"embeddings.":[189],"show":[191,224],"these":[193],"lead":[196],"behaviors.":[199],"Notably,":[200],"embedding":[204],"allows":[205],"intuitive":[207],"way":[208],"transferring":[210],"style":[212],"reference":[215],"target":[219],"output.":[221],"Lastly,":[222],"technique":[226],"enables":[228],"eDiff-I's":[229],"\"paint-with-words\"":[230],"capability.":[231],"A":[232],"user":[233],"can":[234],"select":[235],"word":[237],"input":[240],"paint":[243],"it":[244],"canvas":[247],"control":[249],"output,":[251],"very":[254],"handy":[255],"crafting":[257],"desired":[259],"mind.":[262],"The":[263],"project":[264],"page":[265],"available":[267],"at":[268],"https://deepimagination.cc/eDiff-I/":[269]},"counts_by_year":[{"year":2025,"cited_by_count":21},{"year":2024,"cited_by_count":134},{"year":2023,"cited_by_count":67},{"year":2022,"cited_by_count":1}],"updated_date":"2026-03-25T23:56:10.502304","created_date":"2022-11-08T00:00:00"}