{"id":"https://openalex.org/W7148418756","doi":"https://doi.org/10.48550/arxiv.2604.00394","title":"Deep Networks Favor Simple Data","display_name":"Deep Networks Favor Simple Data","publication_year":2026,"publication_date":"2026-04-01","ids":{"openalex":"https://openalex.org/W7148418756","doi":"https://doi.org/10.48550/arxiv.2604.00394"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.00394","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00394","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2604.00394","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5123019438","display_name":"Weyl Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Lu, Weyl","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132805528","display_name":"Chenjie Hao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao, Chenjie","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5132800674","display_name":"Yubei Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yubei","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5123019438"],"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":0.4555000066757202,"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":0.4555000066757202,"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.10339999943971634,"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/T11612","display_name":"Stochastic Gradient Optimization Techniques","score":0.07900000363588333,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/estimator","display_name":"Estimator","score":0.640999972820282},{"id":"https://openalex.org/keywords/range","display_name":"Range (aeronautics)","score":0.559499979019165},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.4641000032424927},{"id":"https://openalex.org/keywords/rank","display_name":"Rank (graph theory)","score":0.44119998812675476},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.40799999237060547},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.3873000144958496},{"id":"https://openalex.org/keywords/simple","display_name":"Simple (philosophy)","score":0.3792000114917755},{"id":"https://openalex.org/keywords/outlier","display_name":"Outlier","score":0.3774999976158142}],"concepts":[{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.640999972820282},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.559499979019165},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.47609999775886536},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.4641000032424927},{"id":"https://openalex.org/C164226766","wikidata":"https://www.wikidata.org/wiki/Q7293202","display_name":"Rank (graph theory)","level":2,"score":0.44119998812675476},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.40799999237060547},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3968999981880188},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.3873000144958496},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.38350000977516174},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.3792000114917755},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.3774999976158142},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.36059999465942383},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.3555999994277954},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.34709998965263367},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3447999954223633},{"id":"https://openalex.org/C58489278","wikidata":"https://www.wikidata.org/wiki/Q1172284","display_name":"Data set","level":2,"score":0.323199987411499},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3230000138282776},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.2897999882698059},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C739882","wikidata":"https://www.wikidata.org/wiki/Q3560506","display_name":"Anomaly detection","level":2,"score":0.27790001034736633},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.271699994802475},{"id":"https://openalex.org/C189508267","wikidata":"https://www.wikidata.org/wiki/Q17088227","display_name":"Density estimation","level":3,"score":0.2680000066757202},{"id":"https://openalex.org/C117220453","wikidata":"https://www.wikidata.org/wiki/Q5172842","display_name":"Correlation","level":2,"score":0.2623000144958496},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.25429999828338623},{"id":"https://openalex.org/C150921843","wikidata":"https://www.wikidata.org/wiki/Q1170431","display_name":"Resampling","level":2,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.00394","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00394","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2604.00394","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.00394","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"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":{"Estimated":[0],"density":[1,25,71,89,264],"is":[2,11,236],"often":[3],"interpreted":[4],"as":[5,41,153,211],"indicating":[6],"how":[7],"typical":[8],"a":[9,13,52,93,102,145,199,224],"sample":[10,202],"under":[12],"model.":[14],"Yet":[15],"deep":[16,228],"models":[17,180,206],"trained":[18,67,163,188],"on":[19,190],"one":[20],"dataset":[21],"can":[22],"assign":[23],"higher":[24,131,212],"to":[26,32,38,92,101,223,238,243],"simpler":[27,209],"out-of-distribution":[28],"(OOD)":[29],"data":[30],"than":[31],"in-distribution":[33],"test":[34,146],"data.":[35,233],"We":[36,63,79,250],"refer":[37],"this":[39,49,99,240],"behavior":[40],"the":[42,66,70,118,125,191,204,219],"OOD":[43,126,150,221],"anomaly.":[44],"Prior":[45],"work":[46],"typically":[47],"studies":[48],"phenomenon":[50],"within":[51,144],"single":[53,200],"architecture,":[54],"detector,":[55],"or":[56,77,197],"benchmark,":[57],"implicitly":[58],"assuming":[59],"certain":[60],"canonical":[61],"densities.":[62],"instead":[64],"separate":[65],"network":[68],"from":[69,74],"estimator":[72],"built":[73],"its":[75,252],"representations":[76],"outputs.":[78],"introduce":[80,170],"two":[81],"estimators:":[82],"Jacobian-based":[83],"estimators":[84],"and":[85,114,148,155,157,174,181,245,255,263],"autoregressive":[86],"self-estimators,":[87],"making":[88],"analysis":[90],"applicable":[91],"wide":[94],"range":[95,103],"of":[96,104],"models.":[97,164],"Applying":[98],"perspective":[100],"models,":[105,112],"including":[106],"iGPT,":[107],"PixelCNN++,":[108],"Glow,":[109],"score-based":[110],"diffusion":[111],"DINOv2,":[113],"I-JEPA,":[115],"we":[116,169],"find":[117,175],"same":[119],"striking":[120,176],"regularity":[121],"that":[122,257],"goes":[123],"beyond":[124,218],"anomaly:":[127],"lower-complexity":[128],"samples":[129,136,195],"receive":[130,137],"estimated":[132,139],"density,":[133],"while":[134],"higher-complexity":[135],"lower":[138],"density.":[140,213],"This":[141],"ordering":[142],"appears":[143,259],"set":[147],"across":[149,161,179,260],"pairs":[151],"such":[152,201],"CIFAR-10":[154],"SVHN,":[156],"remains":[158],"highly":[159],"consistent":[160],"independently":[162],"To":[165],"quantify":[166],"these":[167],"orderings,":[168],"Spearman":[171],"rank":[172,208],"correlation":[173],"agreement":[177],"both":[178],"with":[182],"external":[183],"complexity":[184],"metrics.":[185],"Even":[186],"when":[187],"only":[189],"lowest-density":[192],"(most":[193],"complex)":[194],"-":[196,203],"even":[198],"resulting":[205],"still":[207],"images":[210],"These":[214],"observations":[215],"lead":[216],"us":[217],"original":[220],"anomaly":[222],"more":[225,248],"general":[226],"conclusion:":[227],"networks":[229],"consistently":[230],"favor":[231],"simple":[232],"Our":[234],"goal":[235],"not":[237],"close":[239],"question,":[241],"but":[242],"define":[244],"visualize":[246],"it":[247,258],"clearly.":[249],"broaden":[251],"empirical":[253],"scope":[254],"show":[256],"architectures,":[261],"objectives,":[262],"estimators.":[265]},"counts_by_year":[],"updated_date":"2026-04-04T06:10:10.580331","created_date":"2026-04-03T00:00:00"}