{"id":"https://openalex.org/W7165009339","doi":"https://doi.org/10.48550/arxiv.2606.17979","title":"STAR: SpatioTemporal Adaptive Reward Allocation for Text-to-Image RL Post-Training","display_name":"STAR: SpatioTemporal Adaptive Reward Allocation for Text-to-Image RL Post-Training","publication_year":2026,"publication_date":"2026-06-16","ids":{"openalex":"https://openalex.org/W7165009339","doi":"https://doi.org/10.48550/arxiv.2606.17979"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.17979","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.17979","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":"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.2606.17979","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138805167","display_name":"Jinjie Shen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shen, Jinjie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138828641","display_name":"Wei Deng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Deng, Wei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138782737","display_name":"Xian Hu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hu, Xian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138803329","display_name":"Daiguo Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Daiguo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138810262","display_name":"Jian Luan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luan, Jian","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":0.6301000118255615,"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":0.6301000118255615,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.18199999630451202,"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/T10028","display_name":"Topic Modeling","score":0.01940000057220459,"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/granularity","display_name":"Granularity","score":0.7077000141143799},{"id":"https://openalex.org/keywords/generative-grammar","display_name":"Generative grammar","score":0.6703000068664551},{"id":"https://openalex.org/keywords/generative-model","display_name":"Generative model","score":0.5952000021934509},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.49639999866485596},{"id":"https://openalex.org/keywords/noise-reduction","display_name":"Noise reduction","score":0.3224000036716461},{"id":"https://openalex.org/keywords/core","display_name":"Core (optical fiber)","score":0.319599986076355}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.777999997138977},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.7077000141143799},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.6703000068664551},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.5952000021934509},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.49639999866485596},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4659000039100647},{"id":"https://openalex.org/C163294075","wikidata":"https://www.wikidata.org/wiki/Q581861","display_name":"Noise reduction","level":2,"score":0.3224000036716461},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.319599986076355},{"id":"https://openalex.org/C2985684807","wikidata":"https://www.wikidata.org/wiki/Q1513879","display_name":"Text generation","level":2,"score":0.3084000051021576},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3050999939441681},{"id":"https://openalex.org/C42058472","wikidata":"https://www.wikidata.org/wiki/Q810214","display_name":"Base (topology)","level":2,"score":0.2847000062465668},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2791000008583069},{"id":"https://openalex.org/C2781249084","wikidata":"https://www.wikidata.org/wiki/Q908656","display_name":"Preference","level":2,"score":0.25}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.17979","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.17979","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":"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.2606.17979","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.17979","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":"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":{"Existing":[0],"RL":[1,95],"post-training":[2,96],"methods":[3],"for":[4,43,69,94],"text-to-image":[5,30,98],"generation":[6,31,45],"usually":[7],"convert":[8],"the":[9,21,25,48,61,75,81,108,114,118,124,141,178,208],"final-image":[10],"reward":[11,210],"into":[12],"a":[13,166],"single":[14],"scalar":[15],"advantage":[16,144],"and":[17,35,47,100,111,137,139,181,190,203,215,220],"apply":[18],"it":[19,67],"with":[20,150],"same":[22,142],"strength":[23],"to":[24,72,145,162],"entire":[26],"generative":[27,76,109],"trajectory.":[28],"However,":[29],"naturally":[32],"has":[33],"temporal":[34],"spatial":[36,128],"structure:":[37],"different":[38,44],"denoising":[39,135],"steps":[40,136],"are":[41],"responsible":[42],"stages,":[46],"content":[49,116],"that":[50,78,117,131,195],"truly":[51,120],"determines":[52],"text":[53,188,201],"alignment":[54],"often":[55],"appears":[56],"only":[57],"in":[58,123],"part":[59],"of":[60,97],"image.":[62],"This":[63],"granularity":[64],"mismatch":[65],"makes":[66],"difficult":[68],"policy":[70,160,169],"updates":[71,161],"focus":[73],"on":[74,183,217],"components":[77],"actually":[79],"affect":[80],"reward.":[82],"To":[83],"address":[84],"this":[85],"issue,":[86],"we":[87],"propose":[88],"\\textbf{SpatioTemporal":[89],"Adaptive":[90],"Reward":[91],"(STAR)":[92],"Allocation}":[93],"diffusion":[99],"flow":[101],"models.":[102],"STAR":[103,156,196],"uses":[104],"text-image":[105],"attention":[106],"inside":[107],"model":[110,180],"starts":[112],"from":[113],"core":[115],"user":[119],"cares":[121],"about":[122],"prompt.":[125],"It":[126],"constructs":[127],"allocation":[129],"maps":[130],"dynamically":[132],"vary":[133],"across":[134],"rollouts,":[138],"allocates":[140],"group-relative":[143],"more":[146],"relevant":[147],"latent":[148],"regions":[149,164],"almost":[151],"no":[152],"additional":[153],"computational":[154],"overhead.":[155],"then":[157],"applies":[158],"stronger":[159],"these":[163],"through":[165],"spatially":[167],"resolved":[168],"objective.":[170],"We":[171],"use":[172],"Stable":[173],"Diffusion":[174],"3.5":[175],"Medium":[176],"as":[177],"base":[179],"evaluate":[182],"three":[184],"tasks:":[185],"GenEval,":[186,218],"OCR":[187],"rendering,":[189,202],"PickScore.":[191],"Experimental":[192],"results":[193],"show":[194],"improves":[197],"compositional":[198],"semantic":[199],"alignment,":[200],"preference":[204],"optimization":[205],"without":[206],"changing":[207],"external":[209],"source,":[211],"achieving":[212],"$\\mathbf{0.9759}$,":[213],"$\\mathbf{0.9757}$,":[214],"$\\mathbf{23.60}$":[216],"OCR,":[219],"PickScore,":[221],"respectively.":[222]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-18T00:00:00"}