{"id":"https://openalex.org/W7157695576","doi":"https://doi.org/10.48550/arxiv.2604.25636","title":"Refinement via Regeneration: Enlarging Modification Space Boosts Image Refinement in Unified Multimodal Models","display_name":"Refinement via Regeneration: Enlarging Modification Space Boosts Image Refinement in Unified Multimodal Models","publication_year":2026,"publication_date":"2026-04-28","ids":{"openalex":"https://openalex.org/W7157695576","doi":"https://doi.org/10.48550/arxiv.2604.25636"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.25636","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.25636","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.2604.25636","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5134853760","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/A5134856954","display_name":"Linqing Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Linqing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134835697","display_name":"Jiangshan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jiangshan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134841879","display_name":"Yang Yue","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yue, Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134860427","display_name":"Zeyu Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Zeyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134847239","display_name":"Zhiyuan Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Zhiyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134867956","display_name":"Qinglin Lu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lu, Qinglin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134857633","display_name":"Gao Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Gao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134870685","display_name":"Chunyu Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Chunyu","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":0.9125999808311462,"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":0.9125999808311462,"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":0.04089999943971634,"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":0.006300000008195639,"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/image-editing","display_name":"Image editing","score":0.6574000120162964},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.555400013923645},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.5231999754905701},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.5108000040054321},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.499099999666214},{"id":"https://openalex.org/keywords/structured-prediction","display_name":"Structured prediction","score":0.4708000123500824}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8273000121116638},{"id":"https://openalex.org/C2776674983","wikidata":"https://www.wikidata.org/wiki/Q545981","display_name":"Image editing","level":3,"score":0.6574000120162964},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.555400013923645},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.5231999754905701},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5184999704360962},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.5108000040054321},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.499099999666214},{"id":"https://openalex.org/C22367795","wikidata":"https://www.wikidata.org/wiki/Q7625208","display_name":"Structured prediction","level":2,"score":0.4708000123500824},{"id":"https://openalex.org/C2989087649","wikidata":"https://www.wikidata.org/wiki/Q176953","display_name":"Image synthesis","level":3,"score":0.3894999921321869},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.3732999861240387},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.3578000068664551},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.2727999985218048},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.2718999981880188},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C2987933465","wikidata":"https://www.wikidata.org/wiki/Q141130","display_name":"Image manipulation","level":3,"score":0.25529998540878296}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.25636","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.25636","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.2604.25636","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.25636","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Unified":[0],"multimodal":[1],"models":[2],"(UMMs)":[3],"integrate":[4],"visual":[5],"understanding":[6],"and":[7,115,128,164],"generation":[8],"within":[9],"a":[10,41,96,142],"single":[11],"framework.":[12],"For":[13],"text-to-image":[14],"(T2I)":[15],"tasks,":[16],"this":[17],"unified":[18],"capability":[19],"allows":[20],"UMMs":[21,46],"to":[22,50,68,157,162,168],"refine":[23],"outputs":[24],"after":[25],"their":[26],"initial":[27,134],"generation,":[28],"potentially":[29],"extending":[30],"the":[31,80,125,129,133,149],"performance":[32],"upper":[33],"bound.":[34],"Current":[35],"UMM-based":[36],"refinement":[37,101],"methods":[38],"primarily":[39],"follow":[40],"refinement-via-editing":[42],"(RvE)":[43],"paradigm,":[44],"where":[45],"produce":[47],"editing":[48,59,113],"instructions":[49,60,114],"modify":[51],"misaligned":[52],"regions":[53],"while":[54],"preserving":[55],"aligned":[56],"content.":[57],"However,":[58],"often":[61],"describe":[62],"prompt-image":[63],"misalignment":[64],"only":[65],"coarsely,":[66],"leading":[67],"incomplete":[69],"refinement.":[70,85],"Moreover,":[71],"pixel-level":[72],"preservation,":[73,119],"though":[74],"necessary":[75],"for":[76,84],"editing,":[77],"unnecessarily":[78],"restricts":[79],"effective":[81],"modification":[82,144],"space":[83],"To":[86],"address":[87],"these":[88],"limitations,":[89],"we":[90],"propose":[91],"Refinement":[92],"via":[93],"Regeneration":[94],"(RvR),":[95],"novel":[97],"framework":[98],"that":[99],"reformulates":[100],"as":[102],"conditional":[103],"image":[104],"regeneration":[105],"rather":[106],"than":[107],"editing.":[108],"Instead":[109],"of":[110,132,151],"relying":[111],"on":[112,124],"enforcing":[116],"strict":[117],"content":[118],"RvR":[120],"regenerates":[121],"images":[122],"conditioned":[123],"target":[126],"prompt":[127],"semantic":[130,139],"tokens":[131],"image,":[135],"enabling":[136],"more":[137],"complete":[138],"alignment":[140],"with":[141],"larger":[143],"space.":[145],"Extensive":[146],"experiments":[147],"demonstrate":[148],"effectiveness":[150],"RvR,":[152],"improving":[153],"Geneval":[154],"from":[155,160,166],"0.78":[156],"0.91,":[158],"DPGBench":[159],"84.02":[161],"87.21,":[163],"UniGenBench++":[165],"61.53":[167],"77.41.":[169]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-30T00:00:00"}