{"id":"https://openalex.org/W7134826599","doi":"https://doi.org/10.48550/arxiv.2603.06993","title":"AdaGen: Learning Adaptive Policy for Image Synthesis","display_name":"AdaGen: Learning Adaptive Policy for Image Synthesis","publication_year":2026,"publication_date":"2026-03-07","ids":{"openalex":"https://openalex.org/W7134826599","doi":"https://doi.org/10.48550/arxiv.2603.06993"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.06993","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.06993","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.2603.06993","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5019968848","display_name":"Zanlin Ni","orcid":"https://orcid.org/0009-0006-6428-1456"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ni, Zanlin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128660314","display_name":"Yulin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yulin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102865400","display_name":"Yue Hua","orcid":"https://orcid.org/0000-0002-1966-5868"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hua, Yeguo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100595910","display_name":"Renping Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Renping","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128670071","display_name":"Jiayi Guo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Jiayi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128634235","display_name":"Jun Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Jun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128666089","display_name":"Bo Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Bo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128673801","display_name":"Gao Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Gao","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":0.796999990940094,"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":0.796999990940094,"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/T11349","display_name":"Music Technology and Sound Studies","score":0.02930000051856041,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.019999999552965164,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.670199990272522},{"id":"https://openalex.org/keywords/markov-decision-process","display_name":"Markov decision process","score":0.5590999722480774},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5024999976158142},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.492900013923645},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.4837000072002411},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.42829999327659607},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.34599998593330383},{"id":"https://openalex.org/keywords/autoregressive-model","display_name":"Autoregressive model","score":0.33809998631477356}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.777899980545044},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.670199990272522},{"id":"https://openalex.org/C106189395","wikidata":"https://www.wikidata.org/wiki/Q176789","display_name":"Markov decision process","level":3,"score":0.5590999722480774},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5098999738693237},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5024999976158142},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.492900013923645},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.4837000072002411},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.47119998931884766},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.42829999327659607},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.34599998593330383},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.33809998631477356},{"id":"https://openalex.org/C37736160","wikidata":"https://www.wikidata.org/wiki/Q1801315","display_name":"Adversarial system","level":2,"score":0.3188000023365021},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3100000023841858},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.30250000953674316},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.29910001158714294},{"id":"https://openalex.org/C23224414","wikidata":"https://www.wikidata.org/wiki/Q176769","display_name":"Hidden Markov model","level":2,"score":0.2816999852657318},{"id":"https://openalex.org/C98763669","wikidata":"https://www.wikidata.org/wiki/Q176645","display_name":"Markov chain","level":2,"score":0.27079999446868896},{"id":"https://openalex.org/C55416958","wikidata":"https://www.wikidata.org/wiki/Q6206757","display_name":"Job shop scheduling","level":3,"score":0.2662999927997589},{"id":"https://openalex.org/C17098449","wikidata":"https://www.wikidata.org/wiki/Q176814","display_name":"Partially observable Markov decision process","level":4,"score":0.26499998569488525},{"id":"https://openalex.org/C2780992000","wikidata":"https://www.wikidata.org/wiki/Q17016113","display_name":"Generator (circuit theory)","level":3,"score":0.2630000114440918}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.06993","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.06993","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.2603.06993","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.06993","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":[{"score":0.6042464375495911,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recent":[0],"advances":[1],"in":[2],"image":[3],"synthesis":[4,36],"have":[5],"been":[6],"propelled":[7],"by":[8],"powerful":[9],"generative":[10,213],"models,":[11,19,21,154],"such":[12,148],"as":[13,116,149],"Masked":[14],"Generative":[15],"Transformers":[16],"(MaskGIT),":[17],"autoregressive":[18],"diffusion":[20],"and":[22,70,101,134,159,194,205,234],"rectified":[23],"flow":[24],"models.":[25],"A":[26],"common":[27],"principle":[28],"behind":[29],"their":[30],"success":[31],"is":[32],"the":[33,77,82,106,113,130,164,181,184,203,216,236],"decomposition":[34],"of":[35,45,85,169,183,207,218,238],"into":[37],"multiple":[38],"steps.":[39],"However,":[40],"this":[41,65,93],"introduces":[42],"a":[43,98,117,122,195],"proliferation":[44],"step-specific":[46],"parameters":[47,128],"(e.g.,":[48],"noise":[49],"level":[50],"or":[51,151,167],"temperature":[52],"at":[53],"each":[54,86],"step).":[55],"Existing":[56],"approaches":[57],"typically":[58],"rely":[59],"on":[60,211,226],"manually-designed":[61],"rules":[62],"to":[63,79,81,179,200,242],"manage":[64],"complexity,":[66],"demanding":[67],"expert":[68],"knowledge":[69],"trial-and-error.":[71],"Furthermore,":[72],"these":[73],"static":[74],"schedules":[75],"lack":[76],"flexibility":[78,206],"adapt":[80],"unique":[83],"characteristics":[84],"sample,":[87],"yielding":[88],"sub-optimal":[89],"performance.":[90],"To":[91],"address":[92],"issue,":[94],"we":[95,111,142,173,188],"present":[96],"AdaGen,":[97],"general,":[99],"learnable,":[100],"sample-adaptive":[102],"framework":[103],"for":[104],"scheduling":[105,114],"iterative":[107],"generation":[108,132],"process.":[109],"Specifically,":[110],"formulate":[112],"problem":[115],"Markov":[118],"Decision":[119],"Process,":[120],"where":[121],"lightweight":[123],"policy":[124,185],"network":[125],"determines":[126],"suitable":[127],"given":[129],"current":[131],"state,":[133],"can":[135,155],"be":[136,156],"trained":[137],"through":[138],"reinforcement":[139],"learning.":[140],"Importantly,":[141],"demonstrate":[143],"that":[144],"simple":[145],"reward":[146,153,177],"designs,":[147],"FID":[150,237],"pre-trained":[152],"easily":[157],"hacked":[158],"may":[160],"not":[161],"reliably":[162],"guarantee":[163],"desired":[165],"quality":[166],"diversity":[168],"generated":[170],"samples.":[171],"Therefore,":[172],"propose":[174],"an":[175,190],"adversarial":[176],"design":[178],"guide":[180],"training":[182],"networks.":[186],"Finally,":[187],"introduce":[189],"inference-time":[191],"refinement":[192],"strategy":[193],"controllable":[196],"fidelity-diversity":[197],"trade-off":[198],"mechanism":[199],"further":[201],"enhance":[202],"performance":[204,225],"AdaGen.":[208,219],"Comprehensive":[209],"experiments":[210],"four":[212],"paradigms":[214],"validate":[215],"superiority":[217],"For":[220],"example,":[221],"AdaGen":[222],"achieves":[223],"better":[224],"DiT-XL":[227],"with":[228,244],"3":[229],"times":[230],"lower":[231],"inference":[232],"cost":[233],"improves":[235],"VAR":[239],"from":[240],"1.92":[241],"1.59":[243],"negligible":[245],"computational":[246],"overhead.":[247]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-03-11T00:00:00"}