{"id":"https://openalex.org/W7166557678","doi":"https://doi.org/10.48550/arxiv.2606.27608","title":"Qwen-Image-2.0-RL Technical Report","display_name":"Qwen-Image-2.0-RL Technical Report","publication_year":2026,"publication_date":"2026-06-25","ids":{"openalex":"https://openalex.org/W7166557678","doi":"https://doi.org/10.48550/arxiv.2606.27608"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.27608","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27608","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.2606.27608","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5027459729","display_name":"Yan Xu","orcid":"https://orcid.org/0000-0003-0379-8866"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Yixian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139588408","display_name":"Kaiyuan Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Kaiyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139568543","display_name":"Yuxiang Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yuxiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139602725","display_name":"Yilei Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yilei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139549986","display_name":"Zecheng Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Zecheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139631378","display_name":"Zihao Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Zihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139586548","display_name":"Zikai Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Zikai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135951014","display_name":"Deqing Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Deqing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139622690","display_name":"Hao Meng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Meng, Hao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135922623","display_name":"Kuan Cao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cao, Kuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139623990","display_name":"Jiahao Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jiahao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139552208","display_name":"Jie Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Jie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139562409","display_name":"Liang Peng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Peng, Liang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109646328","display_name":"Lihan Jiang","orcid":"https://orcid.org/0009-0001-2899-273X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiang, Lihan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135909706","display_name":"Ningyuan Tang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tang, Ningyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139582829","display_name":"Shengming Yin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yin, Shengming","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5063972231","display_name":"Tianhe Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Tianhe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139604967","display_name":"Xiaoyue Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Xiaoyue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139596437","display_name":"Yan Shu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shu, Yan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5071604321","display_name":"Yanran Zhang","orcid":"https://orcid.org/0000-0001-7001-8079"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yanran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139586796","display_name":"Yi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139560698","display_name":"Yu Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139584050","display_name":"Yujia Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yujia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139611509","display_name":"Zekai Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zekai","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139550414","display_name":"Zhendong Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zhendong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139553044","display_name":"Xiao Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Xiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139607285","display_name":"Kun Yan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yan, Kun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5103489382","display_name":"C E Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Chenfei","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":0.6498000025749207,"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":0.6498000025749207,"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/T11448","display_name":"Face recognition and analysis","score":0.09049999713897705,"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":0.054999999701976776,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.6385999917984009},{"id":"https://openalex.org/keywords/fidelity","display_name":"Fidelity","score":0.534600019454956},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.39149999618530273},{"id":"https://openalex.org/keywords/pipeline","display_name":"Pipeline (software)","score":0.35420000553131104},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.3456999957561493},{"id":"https://openalex.org/keywords/reward-system","display_name":"Reward system","score":0.3188000023365021},{"id":"https://openalex.org/keywords/pointwise","display_name":"Pointwise","score":0.3147999942302704},{"id":"https://openalex.org/keywords/credibility","display_name":"Credibility","score":0.31139999628067017}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6513000130653381},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6385999917984009},{"id":"https://openalex.org/C2776459999","wikidata":"https://www.wikidata.org/wiki/Q2119376","display_name":"Fidelity","level":2,"score":0.534600019454956},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48069998621940613},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.44830000400543213},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.39149999618530273},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.35420000553131104},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.3456999957561493},{"id":"https://openalex.org/C143661069","wikidata":"https://www.wikidata.org/wiki/Q670713","display_name":"Reward system","level":2,"score":0.3188000023365021},{"id":"https://openalex.org/C2777984123","wikidata":"https://www.wikidata.org/wiki/Q9248237","display_name":"Pointwise","level":2,"score":0.3147999942302704},{"id":"https://openalex.org/C2780224610","wikidata":"https://www.wikidata.org/wiki/Q1530061","display_name":"Credibility","level":2,"score":0.31139999628067017},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.28630000352859497},{"id":"https://openalex.org/C106436119","wikidata":"https://www.wikidata.org/wiki/Q836575","display_name":"Quality assurance","level":3,"score":0.2644999921321869},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.26159998774528503},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.2597000002861023},{"id":"https://openalex.org/C197129107","wikidata":"https://www.wikidata.org/wiki/Q1921621","display_name":"Merge (version control)","level":2,"score":0.25859999656677246},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.25690001249313354},{"id":"https://openalex.org/C127705205","wikidata":"https://www.wikidata.org/wiki/Q5748245","display_name":"Heuristics","level":2,"score":0.2554999887943268},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.25519999861717224}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.27608","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27608","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.2606.27608","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27608","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":0.5836897492408752,"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"We":[0],"present":[1],"Qwen-Image-2.0-RL,":[2],"a":[3,48,89,96,142],"post-training":[4],"pipeline":[5],"that":[6,153],"applies":[7],"reinforcement":[8],"learning":[9],"from":[10],"human":[11],"feedback":[12],"(RLHF)":[13],"and":[14,24,52,64,78,113,126,174,189],"on-policy":[15,130],"distillation":[16,131],"(OPD)":[17],"to":[18,102],"improve":[19],"both":[20],"the":[21,28,58,72,120,133,163],"visual":[22],"quality":[23],"instruction-following":[25,76],"capability":[26],"of":[27,168],"Qwen-Image-2.0":[29],"diffusion":[30],"model.":[31],"To":[32,118],"provide":[33],"reliable":[34],"reward":[35,41,59,73,85,110,115],"signals,":[36],"we":[37,87,128],"construct":[38],"task-specific":[39],"composite":[40],"models":[42,46,60],"by":[43],"fine-tuning":[44],"vision-language":[45],"with":[47],"pointwise":[49],"scoring":[50],"paradigm":[51],"chain-of-thought":[53],"reasoning.":[54],"For":[55,68],"text-to-image":[56,171],"generation,":[57],"cover":[61],"alignment,":[62],"aesthetics,":[63],"portrait":[65],"fidelity":[66],"dimensions.":[67],"image":[69,177],"editing":[70,190],"tasks,":[71],"system":[74],"addresses":[75],"accuracy":[77],"face":[79],"identity":[80],"preservation.":[81],"Building":[82],"on":[83,159],"this":[84],"system,":[86],"develop":[88],"scalable":[90],"GRPO-based":[91],"RL":[92,122],"training":[93,135],"framework,":[94],"incorporating":[95],"hybrid":[97],"classifier-free":[98],"guidance":[99],"(CFG)":[100],"strategy":[101],"preserve":[103],"pre-trained":[104],"knowledge,":[105],"prompt":[106,187],"curation":[107],"via":[108],"intra-group":[109],"range":[111],"filtering,":[112],"per-category":[114],"weight":[116],"calibration.":[117],"merge":[119],"task-specialized":[121],"policies":[123],"for":[124],"T2I":[125],"editing,":[127],"propose":[129],"as":[132],"final":[134],"stage,":[136],"which":[137],"consolidates":[138],"multiple":[139],"teachers":[140],"into":[141],"single":[143],"student":[144],"model":[145],"through":[146],"trajectory-level":[147],"velocity":[148],"matching.":[149],"Extensive":[150],"evaluation":[151],"shows":[152],"Qwen-Image-2.0-RL":[154],"achieves":[155],"57.84":[156],"overall":[157],"score":[158],"Qwen-Image-Bench":[160],"(+2.61":[161],"over":[162],"base":[164],"model),":[165],"Elo":[166],"ratings":[167],"1193":[169],"in":[170,176,184],"arena":[172,179],"(+78)":[173],"1349":[175],"edit":[178],"(+93),":[180],"demonstrating":[181],"consistent":[182],"gains":[183],"aesthetic":[185],"quality,":[186],"adherence,":[188],"accuracy.":[191]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-30T00:00:00"}