{"id":"https://openalex.org/W7165397863","doi":"https://doi.org/10.48550/arxiv.2606.20536","title":"The FID Lottery: Quantifying Hidden Randomness in Generative-Model Evaluation","display_name":"The FID Lottery: Quantifying Hidden Randomness in Generative-Model Evaluation","publication_year":2026,"publication_date":"2026-06-18","ids":{"openalex":"https://openalex.org/W7165397863","doi":"https://doi.org/10.48550/arxiv.2606.20536"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.20536","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.20536","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.2606.20536","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138974192","display_name":"Nicolas Dufour","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dufour, Nicolas","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138965277","display_name":"Alexei A. Efros","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Efros, Alexei A.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130556908","display_name":"Patrick Perez","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"P\u00e9rez, Patrick","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":0.3425999879837036,"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":0.3425999879837036,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.1712999939918518,"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/T12859","display_name":"Cell Image Analysis Techniques","score":0.13019999861717224,"subfield":{"id":"https://openalex.org/subfields/1304","display_name":"Biophysics"},"field":{"id":"https://openalex.org/fields/13","display_name":"Biochemistry, Genetics and Molecular Biology"},"domain":{"id":"https://openalex.org/domains/1","display_name":"Life Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/randomness","display_name":"Randomness","score":0.7200000286102295},{"id":"https://openalex.org/keywords/measure","display_name":"Measure (data warehouse)","score":0.5525000095367432},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.5077999830245972},{"id":"https://openalex.org/keywords/gaussian","display_name":"Gaussian","score":0.4717000126838684},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.460099995136261},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.4494999945163727},{"id":"https://openalex.org/keywords/variance","display_name":"Variance (accounting)","score":0.3781000077724457},{"id":"https://openalex.org/keywords/variable","display_name":"Variable (mathematics)","score":0.3564000129699707},{"id":"https://openalex.org/keywords/pattern-recognition","display_name":"Pattern recognition (psychology)","score":0.34549999237060547}],"concepts":[{"id":"https://openalex.org/C125112378","wikidata":"https://www.wikidata.org/wiki/Q176640","display_name":"Randomness","level":2,"score":0.7200000286102295},{"id":"https://openalex.org/C2780009758","wikidata":"https://www.wikidata.org/wiki/Q6804172","display_name":"Measure (data warehouse)","level":2,"score":0.5525000095367432},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.5508000254631042},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.5393999814987183},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.5077999830245972},{"id":"https://openalex.org/C163716315","wikidata":"https://www.wikidata.org/wiki/Q901177","display_name":"Gaussian","level":2,"score":0.4717000126838684},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.460099995136261},{"id":"https://openalex.org/C105795698","wikidata":"https://www.wikidata.org/wiki/Q12483","display_name":"Statistics","level":1,"score":0.45170000195503235},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4510999917984009},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.4494999945163727},{"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/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.3781000077724457},{"id":"https://openalex.org/C182365436","wikidata":"https://www.wikidata.org/wiki/Q50701","display_name":"Variable (mathematics)","level":2,"score":0.3564000129699707},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.34549999237060547},{"id":"https://openalex.org/C199540501","wikidata":"https://www.wikidata.org/wiki/Q1520790","display_name":"Bobbin","level":2,"score":0.33739998936653137},{"id":"https://openalex.org/C4199805","wikidata":"https://www.wikidata.org/wiki/Q2725903","display_name":"Gaussian noise","level":2,"score":0.3321000039577484},{"id":"https://openalex.org/C201866948","wikidata":"https://www.wikidata.org/wiki/Q228206","display_name":"Random number generation","level":2,"score":0.3271999955177307},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.30379998683929443},{"id":"https://openalex.org/C18762648","wikidata":"https://www.wikidata.org/wiki/Q42213","display_name":"Work (physics)","level":2,"score":0.29980000853538513},{"id":"https://openalex.org/C122123141","wikidata":"https://www.wikidata.org/wiki/Q176623","display_name":"Random variable","level":2,"score":0.29330000281333923},{"id":"https://openalex.org/C139676723","wikidata":"https://www.wikidata.org/wiki/Q1193832","display_name":"Sign (mathematics)","level":2,"score":0.2793999910354614},{"id":"https://openalex.org/C2780428219","wikidata":"https://www.wikidata.org/wiki/Q16952335","display_name":"Cover (algebra)","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C94966114","wikidata":"https://www.wikidata.org/wiki/Q29256","display_name":"Black box","level":2,"score":0.2784000039100647},{"id":"https://openalex.org/C188721877","wikidata":"https://www.wikidata.org/wiki/Q103510","display_name":"Bar (unit)","level":2,"score":0.27059999108314514},{"id":"https://openalex.org/C42747912","wikidata":"https://www.wikidata.org/wiki/Q1048447","display_name":"Multiplicative function","level":2,"score":0.26649999618530273},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2632000148296356},{"id":"https://openalex.org/C28826006","wikidata":"https://www.wikidata.org/wiki/Q33521","display_name":"Applied mathematics","level":1,"score":0.25859999656677246},{"id":"https://openalex.org/C26263659","wikidata":"https://www.wikidata.org/wiki/Q1886131","display_name":"Random seed","level":3,"score":0.2542000114917755},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.2540999948978424},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.25380000472068787}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.20536","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.20536","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.2606.20536","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.20536","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":{"The":[0],"Frechet":[1],"Inception":[2,101],"Distance":[3],"(FID)":[4],"is":[5,33,114],"the":[6,39,86,89,124,129,140,143,158,172,206],"de":[7],"facto":[8],"arbiter":[9],"of":[10,60,128,146],"image":[11],"generation,":[12],"yet":[13],"most":[14],"papers":[15],"report":[16,81,214],"just":[17],"a":[18,22,27,53,57,93,108,150,167,191,224],"single":[19,23,28,225],"number":[20,35],"from":[21,44,107],"trained":[24,75],"model":[25,87,136],"using":[26,88],"sampling":[29],"seed.":[30],"How":[31],"reproducible":[32],"that":[34],"if":[36],"we":[37,49,189],"retrain":[38],"model,":[40],"or":[41,135],"merely":[42],"resample":[43],"it?":[45],"In":[46],"this":[47],"paper,":[48],"treat":[50,201],"FID":[51,97,144,174,193,203,226],"as":[52,211],"random":[54,119],"variable":[55],"on":[56,70,76,186],"two-axis":[58],"panel":[59],"training":[61,169,220],"and":[62,65,123,166,213],"generation":[63],"seeds,":[64],"measure":[66],"its":[67],"variance":[68],"directly":[69],"several":[71,219],"hundred":[72],"SiT":[73],"networks":[74],"class-conditional":[77],"ImageNet":[78],"256x256.":[79],"We":[80],"surprising":[82],"findings:":[83],"(a)":[84],"Retraining":[85],"same":[90,173],"recipe":[91],"with":[92,175],"different":[94],"seed":[95,170],"moves":[96],"3.2x":[98],"more":[99],"(in":[100],"feature":[102],"space)":[103],"than":[104,181,223],"redrawing":[105],"samples":[106],"fixed":[109],"network.":[110],"(b)":[111],"That":[112],"gap":[113,204],"driven":[115],"by":[116],"three":[117],"factors:":[118],"initialisation,":[120],"data":[121],"ordering,":[122],"per-step":[125],"Gaussian":[126],"noise":[127],"flow-matching":[130],"loss.":[131],"(c)":[132],"Increasing":[133],"compute":[134,180],"size":[137],"barely":[138],"tightens":[139],"spread,":[141],"holding":[142],"coefficient":[145],"variation":[147],"(CoV)":[148],"inside":[149],"1-2%":[151],"band.":[152],"(d)":[153],"Per-cell":[154],"classifier-free-guidance":[155],"tuning":[156],"halves":[157],"spread":[159],"but":[160],"reshuffles":[161],"which":[162],"seeds":[163,221],"work":[164],"best,":[165],"lucky":[168],"reaches":[171],"up":[176],"to":[177],"2x":[178],"less":[179],"an":[182,215],"unlucky":[183],"one.":[184],"Based":[185],"these":[187],"findings,":[188],"recommend":[190],"new":[192],"evaluation":[194],"protocol:":[195],"evaluate":[196],"under":[197],"per-cell":[198],"optimal":[199],"guidance,":[200],"any":[202],"below":[205],"empirically":[207],"measured":[208],"~1.3%":[209],"CoV":[210],"inconclusive,":[212],"error":[216],"bar":[217],"over":[218],"rather":[222],"number.":[227]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-20T00:00:00"}