{"id":"https://openalex.org/W7161741048","doi":"https://doi.org/10.48550/arxiv.2605.17807","title":"Curriculum Group Policy Optimization: Adaptive Sampling for Unleashing the Potential of Text-to-Image Generation","display_name":"Curriculum Group Policy Optimization: Adaptive Sampling for Unleashing the Potential of Text-to-Image Generation","publication_year":2026,"publication_date":"2026-05-18","ids":{"openalex":"https://openalex.org/W7161741048","doi":"https://doi.org/10.48550/arxiv.2605.17807"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.17807","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.17807","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2605.17807","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5112977795","display_name":"Baoteng Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Baoteng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136491421","display_name":"Xianghao Zang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zang, Xianghao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136468274","display_name":"Xinran Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xinran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136482509","display_name":"Xiangyu Na","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Na, Xiangyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136480968","display_name":"Zhixiang He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Zhixiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136460315","display_name":"Hao Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136473356","display_name":"Chi Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Chi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136475297","display_name":"Zhongjiang He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Zhongjiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5072611264","display_name":"Tianwei Cao","orcid":"https://orcid.org/0000-0002-5408-986X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cao, Tianwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136494610","display_name":"Kongming Liang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liang, Kongming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136496734","display_name":"Zhanyu Ma","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Zhanyu","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":0.5236999988555908,"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":0.5236999988555908,"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":0.17659999430179596,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.06639999896287918,"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/variance","display_name":"Variance (accounting)","score":0.5778999924659729},{"id":"https://openalex.org/keywords/curriculum","display_name":"Curriculum","score":0.531499981880188},{"id":"https://openalex.org/keywords/adaptive-sampling","display_name":"Adaptive sampling","score":0.531000018119812},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.5102999806404114},{"id":"https://openalex.org/keywords/sampling","display_name":"Sampling (signal processing)","score":0.5019999742507935},{"id":"https://openalex.org/keywords/sample","display_name":"Sample (material)","score":0.4740000069141388},{"id":"https://openalex.org/keywords/proxy","display_name":"Proxy (statistics)","score":0.46779999136924744},{"id":"https://openalex.org/keywords/training","display_name":"Training (meteorology)","score":0.4020000100135803}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6274999976158142},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.5778999924659729},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.5461999773979187},{"id":"https://openalex.org/C47177190","wikidata":"https://www.wikidata.org/wiki/Q207137","display_name":"Curriculum","level":2,"score":0.531499981880188},{"id":"https://openalex.org/C2781395549","wikidata":"https://www.wikidata.org/wiki/Q4680762","display_name":"Adaptive sampling","level":3,"score":0.531000018119812},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.5102999806404114},{"id":"https://openalex.org/C140779682","wikidata":"https://www.wikidata.org/wiki/Q210868","display_name":"Sampling (signal processing)","level":3,"score":0.5019999742507935},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.4740000069141388},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4700999855995178},{"id":"https://openalex.org/C2780148112","wikidata":"https://www.wikidata.org/wiki/Q1432581","display_name":"Proxy (statistics)","level":2,"score":0.46779999136924744},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.4020000100135803},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.37119999527931213},{"id":"https://openalex.org/C75917345","wikidata":"https://www.wikidata.org/wiki/Q2725298","display_name":"Sampling bias","level":3,"score":0.3628999888896942},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3393000066280365},{"id":"https://openalex.org/C2781311116","wikidata":"https://www.wikidata.org/wiki/Q83306","display_name":"Group (periodic table)","level":2,"score":0.3357999920845032},{"id":"https://openalex.org/C129848803","wikidata":"https://www.wikidata.org/wiki/Q2564360","display_name":"Sample size determination","level":2,"score":0.31679999828338623},{"id":"https://openalex.org/C2779436431","wikidata":"https://www.wikidata.org/wiki/Q30672407","display_name":"Policy learning","level":2,"score":0.2953000068664551},{"id":"https://openalex.org/C100480397","wikidata":"https://www.wikidata.org/wiki/Q7836899","display_name":"Treatment and control groups","level":2,"score":0.2883000075817108},{"id":"https://openalex.org/C155108698","wikidata":"https://www.wikidata.org/wiki/Q1231081","display_name":"Randomized experiment","level":2,"score":0.26080000400543213}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.17807","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.17807","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.17807","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.17807","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":"Preprint"},"sustainable_development_goals":[{"score":0.6105140447616577,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Text-to-Image":[0],"(T2I)":[1],"generation":[2,199],"has":[3,131,138],"achieved":[4,141],"remarkable":[5],"progress":[6],"in":[7,166],"recent":[8],"years.":[9],"Meanwhile,":[10],"reinforcement":[11],"learning":[12,54,152],"methods,":[13],"particularly":[14],"those":[15],"based":[16,175],"on":[17,176,187],"Group":[18,87],"Relative":[19],"Policy":[20,88],"Optimization":[21,89],"(GRPO),":[22],"have":[23],"attracted":[24],"widespread":[25],"attention":[26],"and":[27,50,77,190],"been":[28],"successfully":[29],"applied":[30],"to":[31,57,149,162],"T2I":[32],"tasks.":[33],"However,":[34],"the":[35,45,51,73,112,129,134],"uniform":[36],"sampling":[37,158],"strategy":[38],"commonly":[39],"used":[40],"during":[41],"training":[42,59,65,94,182],"often":[43],"ignores":[44],"match":[46,72],"between":[47],"sample":[48],"difficulty":[49,183],"model's":[52,74],"current":[53],"capability,":[55],"leading":[56],"low":[58],"efficiency.":[60],"We":[61,110],"argue":[62],"that":[63,71,128,194],"improving":[64],"efficiency":[66],"requires":[67],"continuously":[68],"prioritizing":[69],"prompts":[70,145],"evolving":[75],"capability":[76],"remain":[78],"actively":[79],"learnable.":[80],"To":[81],"this":[82],"end,":[83],"we":[84,155,169],"propose":[85],"Curriculum":[86],"(CGPO),":[90],"an":[91,118],"adaptive":[92],"curriculum":[93],"framework.":[95],"During":[96],"training,":[97],"each":[98],"prompt":[99,122,135],"produces":[100],"a":[101,107,171],"group":[102,115],"of":[103,114],"images":[104],"scored":[105],"by":[106],"reward":[108],"model.":[109],"use":[111],"variance":[113,126],"rewards":[116],"as":[117],"online":[119],"proxy":[120],"for":[121],"inconsistency.":[123],"A":[124],"higher":[125],"suggests":[127],"model":[130],"partially":[132],"captured":[133],"requirements":[136],"but":[137],"not":[139],"yet":[140],"stable":[142],"mastery.":[143],"Such":[144],"are":[146],"more":[147],"likely":[148],"provide":[150],"useful":[151],"signals,":[153],"so":[154],"increase":[156],"their":[157],"probabilities":[159],"accordingly.":[160],"Additionally,":[161],"address":[163],"data":[164],"imbalance":[165],"multi-category":[167],"datasets,":[168],"design":[170],"category":[172],"calibration":[173],"method":[174],"proportional":[177],"fairness":[178],"optimization,":[179],"which":[180],"balances":[181],"across":[184],"categories.":[185],"Experiments":[186],"GenEval,":[188],"T2I-CompBench++,":[189],"DPG":[191],"Bench":[192],"demonstrate":[193],"our":[195],"framework":[196],"effectively":[197],"improves":[198],"performance.":[200]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-20T00:00:00"}