{"id":"https://openalex.org/W4404346261","doi":"https://doi.org/10.48550/arxiv.2410.14919","title":"Adversarial Score identity Distillation: Rapidly Surpassing the Teacher in One Step","display_name":"Adversarial Score identity Distillation: Rapidly Surpassing the Teacher in One Step","publication_year":2024,"publication_date":"2024-10-19","ids":{"openalex":"https://openalex.org/W4404346261","doi":"https://doi.org/10.48550/arxiv.2410.14919"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2410.14919","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.14919","pdf_url":"https://arxiv.org/pdf/2410.14919","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"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2410.14919","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5010394308","display_name":"Mingyuan Zhou","orcid":"https://orcid.org/0000-0002-4253-2780","observed_orcids":["https://orcid.org/0000-0002-4253-2780"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Mingyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5077499817","display_name":"Huangjie Zheng","orcid":"https://orcid.org/0000-0003-0508-5034","observed_orcids":["https://orcid.org/0000-0003-0508-5034","https://orcid.org/0009-0004-4910-5679"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Huangjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014319412","display_name":"Yi Gu","orcid":"https://orcid.org/0000-0001-7962-9466","observed_orcids":["https://orcid.org/0000-0001-7962-9466"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gu, Yi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100418326","display_name":"Zhendong Wang","orcid":"https://orcid.org/0000-0002-7082-2478","observed_orcids":["https://orcid.org/0000-0002-7082-2478","https://orcid.org/0009-0007-1917-4691"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zhendong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Huang, Hai","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Hai","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":0.9994999766349792,"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":0.9994999766349792,"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":9.999999747378752e-05,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.0,"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/adversarial-loss","display_name":"adversarial loss","score":0.7979999780654907},{"id":"https://openalex.org/keywords/diffusion-models","display_name":"diffusion models","score":0.6179999709129333},{"id":"https://openalex.org/keywords/batch-normalization","display_name":"batch normalization","score":0.46799999475479126},{"id":"https://openalex.org/keywords/fid-score","display_name":"FID score","score":0.4449999928474426}],"concepts":[{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.8617605566978455},{"id":"https://openalex.org/C2778355321","wikidata":"https://www.wikidata.org/wiki/Q17079427","display_name":"Identity (music)","level":2,"score":0.7329464554786682},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.7067293524742126},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.47634509205818176},{"id":"https://openalex.org/C145420912","wikidata":"https://www.wikidata.org/wiki/Q853077","display_name":"Mathematics education","level":1,"score":0.33496493101119995},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3147735595703125},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.3038782477378845},{"id":"https://openalex.org/C185592680","wikidata":"https://www.wikidata.org/wiki/Q2329","display_name":"Chemistry","level":0,"score":0.23845529556274414},{"id":"https://openalex.org/C43617362","wikidata":"https://www.wikidata.org/wiki/Q170050","display_name":"Chromatography","level":1,"score":0.19176706671714783},{"id":"https://openalex.org/C142362112","wikidata":"https://www.wikidata.org/wiki/Q735","display_name":"Art","level":0,"score":0.16387644410133362},{"id":"https://openalex.org/C107038049","wikidata":"https://www.wikidata.org/wiki/Q35986","display_name":"Aesthetics","level":1,"score":0.07406172156333923}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2410.14919","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.14919","pdf_url":"https://arxiv.org/pdf/2410.14919","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.2410.14919","doi":"https://doi.org/10.48550/arxiv.2410.14919","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2410.14919","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:2410.14919","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2410.14919","pdf_url":"https://arxiv.org/pdf/2410.14919","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":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4394591914","https://openalex.org/W4417302166","https://openalex.org/W4399403247","https://openalex.org/W4415795833","https://openalex.org/W4402706167","https://openalex.org/W7161993011","https://openalex.org/W4409347087","https://openalex.org/W4415797645","https://openalex.org/W4415235512","https://openalex.org/W7083697187"],"abstract_inverted_index":{"Score":[0],"identity":[1],"Distillation":[2],"(SiD)":[3],"is":[4,32,107,271],"a":[5,19,90,139,164,256],"data-free":[6],"method":[7,172,202],"that":[8],"has":[9],"achieved":[10,211],"SOTA":[11],"performance":[12,31,160,178],"in":[13,176,255],"image":[14],"generation":[15,66,177,222,258],"by":[16,34,73,102],"leveraging":[17],"only":[18,64],"pretrained":[20,38],"diffusion":[21,50,182],"model,":[22,208],"without":[23,252],"requiring":[24],"any":[25],"training":[26],"data.":[27],"However,":[28],"its":[29,147],"ultimate":[30],"constrained":[33],"how":[35],"accurate":[36],"the":[37,41,49,82,85,116,124,131,157,205],"model":[39,267],"captures":[40],"true":[42],"data":[43],"scores":[44,186,229],"at":[45,273],"different":[46],"stages":[47],"of":[48,187,214,230],"process.":[51],"In":[52,224],"this":[53],"paper,":[54],"we":[55],"introduce":[56],"SiDA":[57,80,136,142,201,226],"(SiD":[58],"with":[59,115],"Adversarial":[60],"Loss),":[61],"which":[62,210],"not":[63],"enhances":[65],"quality":[67],"but":[68],"also":[69],"improves":[70,155],"distillation":[71,171],"efficiency":[72],"incorporating":[74],"real":[75,97],"images":[76,98],"and":[77,99,112,153,220,247,254],"adversarial":[78,105,170],"loss.":[79,119],"utilizes":[81],"encoder":[83],"from":[84,151,163],"generator's":[86],"score":[87],"network":[88],"as":[89],"discriminator,":[91],"allowing":[92],"it":[93],"to":[94,137],"distinguish":[95],"between":[96],"those":[100],"generated":[101],"SiD.":[103],"The":[104],"loss":[106],"batch-normalized":[108],"within":[109],"each":[110],"GPU":[111,128],"then":[113],"combined":[114],"original":[117,158],"SiD":[118,133,166],"This":[120,168],"integration":[121],"effectively":[122],"incorporates":[123],"average":[125],"\"fakeness\"":[126],"per":[127],"batch":[129],"into":[130],"pixel-based":[132],"loss,":[134],"enabling":[135],"distill":[138],"single-step":[140],"generator.":[141,167],"converges":[143],"significantly":[144],"faster":[145],"than":[146],"predecessor":[148],"when":[149,179],"distilled":[150],"scratch,":[152],"swiftly":[154],"upon":[156],"model's":[159],"during":[161],"fine-tuning":[162],"pre-distilled":[165],"one-step":[169],"establishes":[173],"new":[174],"benchmarks":[175],"distilling":[180,193],"EDM":[181],"models,":[183],"achieving":[184],"FID":[185,213,228],"1.110":[188],"on":[189,197],"ImageNet":[190,198],"64x64.":[191],"When":[192],"EDM2":[194],"models":[195],"trained":[196],"512x512,":[199],"our":[200],"surpasses":[203],"even":[204],"largest":[206],"teacher":[207],"EDM2-XXL,":[209],"an":[212],"1.81":[215],"using":[216],"classifier-free":[217],"guidance":[218],"(CFG)":[219],"63":[221],"steps.":[223],"contrast,":[225],"achieves":[227],"2.156":[231],"for":[232,236,239,242,245,249],"size":[233],"XS,":[234],"1.669":[235],"S,":[237],"1.488":[238],"M,":[240],"1.413":[241],"L,":[243],"1.379":[244],"XL,":[246],"1.366":[248],"XXL,":[250],"all":[251,266],"CFG":[253],"single":[257],"step.":[259],"These":[260],"results":[261],"highlight":[262],"substantial":[263],"improvements":[264],"across":[265],"sizes.":[268],"Our":[269],"code":[270],"available":[272],"https://github.com/mingyuanzhou/SiD/tree/sida.":[274]},"counts_by_year":[],"updated_date":"2026-09-27T08:07:27.782135","created_date":"2024-11-14T00:00:00"}