{"id":"https://openalex.org/W7168353003","doi":"https://doi.org/10.48448/qj78-5842","title":"UniCorn: Towards Self-Improving Unified Multimodal Models through Self-Generated Supervision","display_name":"UniCorn: Towards Self-Improving Unified Multimodal Models through Self-Generated Supervision","publication_year":2026,"publication_date":"2026-06-03","ids":{"openalex":"https://openalex.org/W7168353003","doi":"https://doi.org/10.48448/qj78-5842"},"language":"en","primary_location":{"id":"doi:10.48448/qj78-5842","doi":"https://doi.org/10.48448/qj78-5842","is_oa":true,"landing_page_url":"https://doi.org/10.48448/qj78-5842","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"volume":null,"type":"other","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48448/qj78-5842","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5140798803","display_name":"Association for Computational Linguistics 2026","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Association for Computational Linguistics 2026","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140877355","display_name":"Yi Cao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I3923682","display_name":"Soochow University","ror":"https://ror.org/05t8y2r12","country_code":"CN","type":"education","lineage":["https://openalex.org/I3923682"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Cao, Yi","raw_affiliation_strings":["Soochow University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Soochow University","institution_ids":["https://openalex.org/I3923682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140864373","display_name":"Lin Chen","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Lin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140900593","display_name":"Zehui Chen","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Zehui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125627903","display_name":"Zhen Fang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Zhen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122038722","display_name":"Ruiyan Han","orcid":"https://orcid.org/0009-0008-0988-4750","observed_orcids":["https://orcid.org/0009-0008-0988-4750"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Ruiyan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140781426","display_name":"Wenxuan Huang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Wenxuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140929419","display_name":"Yuchen Ma","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ma, Yuchen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140868223","display_name":"Xinyu Sun","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Xinyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140817121","display_name":"Ziheng Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Ziheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140828871","display_name":"Wei-Jie Xu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Wei-Jie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140830536","display_name":"Yu Zeng","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zeng, Yu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5140850441","display_name":"Feng Zhao","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Feng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9585000276565552,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9585000276565552,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.041099999099969864,"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":0.0,"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/unified-multimodal-models","display_name":"unified multimodal models","score":0.996999979019165},{"id":"https://openalex.org/keywords/personal-development","display_name":"human development","score":0.7609999775886536},{"id":"https://openalex.org/keywords/conduction-aphasia","display_name":"conduction aphasia","score":0.8209999799728394},{"id":"https://openalex.org/keywords/text-to-image-generation","display_name":"text-to-image generation","score":0.8019999861717224},{"id":"https://openalex.org/keywords/self-play","display_name":"self-play","score":0.6549999713897705},{"id":"https://openalex.org/keywords/image-generation","display_name":"image generation","score":0.3790000081062317}],"concepts":[{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.7484999895095825},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7394999861717224},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6134999990463257},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5794000029563904},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.5120999813079834},{"id":"https://openalex.org/C2780734062","wikidata":"https://www.wikidata.org/wiki/Q3549947","display_name":"Unicorn","level":2,"score":0.47679999470710754},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4528000056743622},{"id":"https://openalex.org/C2780440489","wikidata":"https://www.wikidata.org/wiki/Q5227278","display_name":"Data-driven","level":2,"score":0.41760000586509705},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4172999858856201},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.41110000014305115},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.36970001459121704},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.32170000672340393},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.2971999943256378},{"id":"https://openalex.org/C45493050","wikidata":"https://www.wikidata.org/wiki/Q7884934","display_name":"Unified Model","level":2,"score":0.2948000133037567},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.27559998631477356},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.267300009727478}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48448/qj78-5842","doi":"https://doi.org/10.48448/qj78-5842","is_oa":true,"landing_page_url":"https://doi.org/10.48448/qj78-5842","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"}],"best_oa_location":{"id":"doi:10.48448/qj78-5842","doi":"https://doi.org/10.48448/qj78-5842","is_oa":true,"landing_page_url":"https://doi.org/10.48448/qj78-5842","pdf_url":null,"source":{"id":"https://openalex.org/S7407053148","display_name":"Underline Science Inc.","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"listed_in":[],"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"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":"Audiovisual"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4416611615","https://openalex.org/W7166108033","https://openalex.org/W7134237298","https://openalex.org/W7168850937","https://openalex.org/W4407384793","https://openalex.org/W7211999619","https://openalex.org/W4416612959","https://openalex.org/W7154625525","https://openalex.org/W4415272862","https://openalex.org/W7220473235"],"abstract_inverted_index":{"While":[0],"Unified":[1],"Multimodal":[2],"Models":[3],"(UMMs)":[4],"have":[5],"achieved":[6],"remarkable":[7],"success":[8],"in":[9,16],"cross-modal":[10],"comprehension,":[11,187],"a":[12,34,59,77,117,122],"significant":[13],"gap":[14],"persists":[15],"their":[17],"ability":[18],"to":[19,44,99,124,126],"leverage":[20],"such":[21],"internal":[22],"knowledge":[23],"for":[24,69,195],"high-quality":[25,90],"generation.":[26],"We":[27],"formalize":[28],"this":[29],"discrepancy":[30],"as":[31],"Conduction":[32],"Aphasia,":[33],"phenomenon":[35],"where":[36],"models":[37],"accurately":[38],"interpret":[39],"multimodal":[40,112,197],"inputs":[41],"but":[42],"struggle":[43],"translate":[45],"that":[46,65,133,177],"understanding":[47,102],"into":[48,80,103],"faithful":[49],"and":[50,86,94,137,159,170],"controllable":[51],"synthesis.":[52],"To":[53,107],"address":[54],"this,":[55],"we":[56,114],"propose":[57],"UniCorn,":[58],"simple":[60],"yet":[61],"elegant":[62],"self-improvement":[63],"framework":[64],"eliminates":[66],"the":[67,109,141,189],"need":[68],"external":[70],"data":[71],"or":[72],"teacher":[73],"supervision.":[74],"By":[75],"partitioning":[76],"single":[78],"UMM":[79],"three":[81],"collaborative":[82],"roles:":[83],"Proposer,":[84],"Solver,":[85],"Judge,":[87],"UniCorn":[88,134],"generates":[89],"interactions":[91],"via":[92],"self-play":[93],"employs":[95],"cognitive":[96],"pattern":[97],"reconstruction":[98,128],"distill":[100],"latent":[101],"explicit":[104],"generative":[105],"signals.":[106],"validate":[108],"restoration":[110],"of":[111,166,191],"coherence,":[113],"introduce":[115],"UniCycle,":[116],"cycle-consistency":[118],"benchmark":[119],"based":[120],"on":[121,155,168,172],"Text":[123,127],"Image":[125],"loop.":[129],"Extensive":[130],"experiments":[131],"demonstrate":[132],"achieves":[135,152],"comprehensive":[136],"substantial":[138,164],"improvements":[139],"over":[140],"base":[142],"model":[143],"across":[144],"six":[145],"general":[146],"image":[147],"generation":[148,183],"benchmarks.":[149],"Notably,":[150],"it":[151],"SOTA":[153],"performance":[154],"TIIF(73.8),":[156],"DPG(86.8),":[157],"CompBench(88.5),":[158],"UniCycle":[160],"while":[161,184],"further":[162],"delivering":[163],"gains":[165],"+5.0":[167],"WISE":[169],"+6.5":[171],"OneIG.":[173],"These":[174],"results":[175],"highlight":[176],"our":[178],"method":[179],"significantly":[180],"enhances":[181],"T2I":[182],"maintaining":[185],"robust":[186],"demonstrating":[188],"scalability":[190],"fully":[192],"self-supervised":[193],"refinement":[194],"unified":[196],"intelligence.":[198]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-07-16T00:00:00"}