{"id":"https://openalex.org/W7138234206","doi":"https://doi.org/10.1609/aaai.v40i8.37611","title":"MoSs: Mixture of Scales for Efficient High-Resolution Autoregressive Image Generation","display_name":"MoSs: Mixture of Scales for Efficient High-Resolution Autoregressive Image Generation","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138234206","doi":"https://doi.org/10.1609/aaai.v40i8.37611"},"language":"en","primary_location":{"id":"doi:10.1609/aaai.v40i8.37611","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i8.37611","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i8.37611","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5102585583","display_name":"Yaoxiu Lian","orcid":"https://orcid.org/0009-0007-7858-5132"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yaoxiu Lian","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129704259","display_name":"Hao LIANG","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Liang","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129649612","display_name":"Zhihong Gou","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhihong Gou","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129745825","display_name":"Yijia Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yijia Zhang","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129744033","display_name":"Jiaming Xu","orcid":null},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jiaming Xu","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129718543","display_name":"Guohao Dai","orcid":null},"institutions":[{"id":"https://openalex.org/I2802974365","display_name":"Shanghai Innovative Research Center of Traditional Chinese Medicine","ror":"https://ror.org/01gnagj68","country_code":"CN","type":"facility","lineage":["https://openalex.org/I2802974365"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guohao Dai","raw_affiliation_strings":["Shanghai Jiao Tong University\nInfinigence-AI\nShanghai Innovation Institute (SII)"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University\nInfinigence-AI\nShanghai Innovation Institute (SII)","institution_ids":["https://openalex.org/I2802974365"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5018513988","display_name":"Ningyi Xu","orcid":"https://orcid.org/0009-0004-6809-7694"},"institutions":[{"id":"https://openalex.org/I183067930","display_name":"Shanghai Jiao Tong University","ror":"https://ror.org/0220qvk04","country_code":"CN","type":"education","lineage":["https://openalex.org/I183067930"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ningyi Xu","raw_affiliation_strings":["Shanghai Jiao Tong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Shanghai Jiao Tong University","institution_ids":["https://openalex.org/I183067930"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":"40","issue":"8","first_page":"6789","last_page":"6797"},"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":0.8640000224113464,"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":0.8640000224113464,"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/T11019","display_name":"Image Enhancement Techniques","score":0.029899999499320984,"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/T11105","display_name":"Advanced Image Processing Techniques","score":0.025800000876188278,"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/security-token","display_name":"Security token","score":0.7612000107765198},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.692300021648407},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5092999935150146},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.49880000948905945},{"id":"https://openalex.org/keywords/pixel","display_name":"Pixel","score":0.42899999022483826},{"id":"https://openalex.org/keywords/dependency","display_name":"Dependency (UML)","score":0.4205000102519989},{"id":"https://openalex.org/keywords/image-quality","display_name":"Image quality","score":0.3930000066757202},{"id":"https://openalex.org/keywords/image-resolution","display_name":"Image resolution","score":0.3695000112056732},{"id":"https://openalex.org/keywords/quality","display_name":"Quality (philosophy)","score":0.3659999966621399}],"concepts":[{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7612000107765198},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6978999972343445},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.692300021648407},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5092999935150146},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.49880000948905945},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4668999910354614},{"id":"https://openalex.org/C160633673","wikidata":"https://www.wikidata.org/wiki/Q355198","display_name":"Pixel","level":2,"score":0.42899999022483826},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.4205000102519989},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4050999879837036},{"id":"https://openalex.org/C55020928","wikidata":"https://www.wikidata.org/wiki/Q3813865","display_name":"Image quality","level":3,"score":0.3930000066757202},{"id":"https://openalex.org/C205372480","wikidata":"https://www.wikidata.org/wiki/Q210521","display_name":"Image resolution","level":2,"score":0.3695000112056732},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3659999966621399},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.35109999775886536},{"id":"https://openalex.org/C97256817","wikidata":"https://www.wikidata.org/wiki/Q1462316","display_name":"Spurious relationship","level":2,"score":0.3366999924182892},{"id":"https://openalex.org/C195956108","wikidata":"https://www.wikidata.org/wiki/Q7268362","display_name":"Quadratic growth","level":2,"score":0.3305000066757202},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.30140000581741333},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.3000999987125397},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.29670000076293945},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.2928999960422516},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.2694999873638153},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.26499998569488525},{"id":"https://openalex.org/C207609745","wikidata":"https://www.wikidata.org/wiki/Q4944086","display_name":"Bootstrapping (finance)","level":2,"score":0.2612999975681305},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.25440001487731934},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2538999915122986},{"id":"https://openalex.org/C2776502983","wikidata":"https://www.wikidata.org/wiki/Q690182","display_name":"Contrast (vision)","level":2,"score":0.2531999945640564},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.25200000405311584}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1609/aaai.v40i8.37611","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i8.37611","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},{"id":"pmh:oai:ojs.aaai.org:article/37611","is_oa":false,"landing_page_url":"https://ojs.aaai.org/index.php/AAAI/article/view/37611","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":"2159-5399","raw_type":"info:eu-repo/semantics/article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i8.37611","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i8.37611","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"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":{"Since":[0],"next-scale":[1,27],"prediction":[2,28],"was":[3],"introduced":[4],"as":[5],"a":[6,24,45,59,121,156],"new":[7],"paradigm":[8,43],"for":[9,128,155],"autoregressive":[10,131],"image":[11,72,86,132,169],"generation,":[12],"it":[13],"has":[14],"attracted":[15],"extensive":[16],"research":[17],"interest.":[18],"By":[19],"progressively":[20],"increasing":[21],"resolution":[22],"in":[23,32,58,142],"draft-to-refinement":[25],"process,":[26],"demonstrates":[29],"great":[30],"potential":[31],"both":[33],"generation":[34,201],"quality":[35,113,202],"and":[36,52,95,170],"efficiency.":[37],"However,":[38],"at":[39],"high":[40],"resolutions,":[41,152],"this":[42],"faces":[44],"fundamental":[46],"challenge:":[47],"token":[48,100,164],"sequences":[49],"grow":[50],"quadratically":[51],"accumulate":[53],"across":[54,83,140],"multiple":[55],"scales,":[56],"resulting":[57],"key":[60],"performance":[61],"bottleneck.":[62],"Our":[63],"systematic":[64],"study":[65],"uncovers":[66],"two":[67],"critical":[68],"observations:":[69],"(1)":[70],"most":[71],"regions":[73,172],"have":[74],"stabilized":[75],"during":[76],"early":[77],"drafting":[78],"stages,":[79],"making":[80],"later":[81],"refinement":[82,145],"the":[84,111,136,143,167,174,183,187,214,232],"full-scale":[85],"token-inefficient;":[87],"(2)":[88],"different":[89,103,151],"scales":[90,104,141,149],"inherently":[91],"trade":[92],"off":[93],"efficiency":[94,196],"fidelity,":[96],"suggesting":[97],"that":[98],"adaptive":[99],"dispatch":[101],"on":[102,200,203,213,227,231],"can":[105],"focus":[106],"resources":[107],"where":[108],"they":[109],"yield":[110],"greatest":[112],"gains.":[114],"Motivated":[115],"by":[116,147],"these":[117],"insights,":[118],"we":[119],"propose":[120],"training-free":[122],"Mixture":[123],"of":[124,150,158],"Scales":[125],"(MoSs)":[126],"method":[127,193],"efficient":[129],"high-resolution":[130,189],"generation.":[133],"MoSs":[134],"breaks":[135],"strict":[137],"causal":[138],"dependency":[139],"final":[144,188],"steps":[146],"parallelizing":[148],"each":[153],"responsible":[154],"subset":[157],"spatial":[159],"regions.":[160],"A":[161],"lightweight":[162],"frequency-based":[163],"dispatcher":[165],"analyzes":[166],"drafted":[168],"assigns":[171],"to":[173,185,218],"appropriate":[175],"scale.":[176],"The":[177,191],"outputs":[178],"are":[179],"then":[180],"composited":[181],"over":[182],"draft":[184],"produce":[186],"image.":[190],"scale-mixture":[192],"exhibits":[194],"remarkable":[195],"with":[197],"little":[198],"impact":[199],"various":[204],"models.":[205],"For":[206],"instance,":[207],"our":[208],"implementation":[209],"achieves":[210],"2.05-4.96x":[211],"speedup":[212],"transformer":[215],"backbone,":[216],"up":[217],"85.62%":[219],"KV":[220],"cache":[221],"reduction,":[222],"incurring":[223],"only":[224],"0.1-2.4%":[225],"loss":[226],"GenEval":[228],"quality,":[229],"based":[230],"state-of-the-art":[233],"Infinity":[234],"model.":[235]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-03-18T00:00:00"}