{"id":"https://openalex.org/W7134966977","doi":"https://doi.org/10.48550/arxiv.2603.09206","title":"MM-Zero: Self-Evolving Multi-Model Vision Language Models From Zero Data","display_name":"MM-Zero: Self-Evolving Multi-Model Vision Language Models From Zero Data","publication_year":2026,"publication_date":"2026-03-10","ids":{"openalex":"https://openalex.org/W7134966977","doi":"https://doi.org/10.48550/arxiv.2603.09206"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.09206","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.09206","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2603.09206","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5128754684","display_name":"Zongxia Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Zongxia","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128695781","display_name":"Hongyang Du","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Du, Hongyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128711571","display_name":"Chengsong Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Chengsong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078545561","display_name":"Xiyang Wu","orcid":"https://orcid.org/0000-0001-8538-8267"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xiyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068559249","display_name":"Lantao Yu","orcid":"https://orcid.org/0000-0003-0569-952X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Lantao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128729015","display_name":"Yicheng He","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"He, Yicheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128692080","display_name":"Jing Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Jing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128756070","display_name":"Xiaomin Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Xiaomin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128736918","display_name":"Zhichao Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Zhichao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5128800325","display_name":"Jiarui Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Jiarui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5128715708","display_name":"Fuxiao Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Fuxiao","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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8701000213623047,"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":0.8701000213623047,"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":0.01730000041425228,"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"}},{"id":"https://openalex.org/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.016499999910593033,"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/executable","display_name":"Executable","score":0.5898000001907349},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5667999982833862},{"id":"https://openalex.org/keywords/visual-reasoning","display_name":"Visual reasoning","score":0.5572999715805054},{"id":"https://openalex.org/keywords/solver","display_name":"Solver","score":0.45320001244544983},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.4368000030517578},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.40560001134872437},{"id":"https://openalex.org/keywords/language-understanding","display_name":"Language understanding","score":0.4018999934196472},{"id":"https://openalex.org/keywords/path","display_name":"Path (computing)","score":0.3937999904155731}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.788100004196167},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5956000089645386},{"id":"https://openalex.org/C160145156","wikidata":"https://www.wikidata.org/wiki/Q778586","display_name":"Executable","level":2,"score":0.5898000001907349},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5667999982833862},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.5572999715805054},{"id":"https://openalex.org/C2778770139","wikidata":"https://www.wikidata.org/wiki/Q1966904","display_name":"Solver","level":2,"score":0.45320001244544983},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.4368000030517578},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.40560001134872437},{"id":"https://openalex.org/C2983448237","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Language understanding","level":2,"score":0.4018999934196472},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.3937999904155731},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3847000002861023},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.36570000648498535},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3605000078678131},{"id":"https://openalex.org/C204323151","wikidata":"https://www.wikidata.org/wiki/Q905424","display_name":"Range (aeronautics)","level":2,"score":0.33570000529289246},{"id":"https://openalex.org/C2780226545","wikidata":"https://www.wikidata.org/wiki/Q6888030","display_name":"Modality (human\u2013computer interaction)","level":2,"score":0.32919999957084656},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.328900009393692},{"id":"https://openalex.org/C2780878386","wikidata":"https://www.wikidata.org/wiki/Q1659648","display_name":"Visual language","level":2,"score":0.32010000944137573},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.3197000026702881},{"id":"https://openalex.org/C164155591","wikidata":"https://www.wikidata.org/wiki/Q2067766","display_name":"Satisfiability modulo theories","level":2,"score":0.3109999895095825},{"id":"https://openalex.org/C2781235140","wikidata":"https://www.wikidata.org/wiki/Q275131","display_name":"Scratch","level":2,"score":0.3093999922275543},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.3028999865055084},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.29589998722076416},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.28369998931884766},{"id":"https://openalex.org/C16345878","wikidata":"https://www.wikidata.org/wiki/Q107472979","display_name":"Orientation (vector space)","level":2,"score":0.27730000019073486},{"id":"https://openalex.org/C5339829","wikidata":"https://www.wikidata.org/wiki/Q1425977","display_name":"Machine vision","level":2,"score":0.272599995136261},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.26930001378059387},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.2612000107765198}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.09206","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.09206","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2603.09206","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.09206","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":[{"id":"https://metadata.un.org/sdg/16","display_name":"Peace, Justice and strong institutions","score":0.42450520396232605}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Self-evolving":[0],"has":[1],"emerged":[2],"as":[3,12,58],"a":[4,95,104,114,131,184,192],"key":[5],"paradigm":[6],"for":[7,82,199],"improving":[8],"foundational":[9],"models":[10],"such":[11,57],"Large":[13],"Language":[14,19],"Models":[15,20],"(LLMs)":[16],"and":[17,90,111,130,152,171],"Vision":[18],"(VLMs)":[21],"with":[22,38,160],"minimal":[23],"human":[24],"intervention.":[25],"While":[26],"recent":[27],"approaches":[28],"have":[29],"demonstrated":[30],"that":[31,49,106,116,133,165,177],"LLM":[32],"agents":[33],"can":[34],"self-evolve":[35],"from":[36,147],"scratch":[37],"little":[39],"to":[40,60,78,126],"no":[41],"data,":[42,56],"VLMs":[43],"introduce":[44],"an":[45],"additional":[46],"visual":[47,109,128,140,169],"modality":[48],"typically":[50],"requires":[51],"at":[52],"least":[53],"some":[54],"seed":[55],"images,":[59],"bootstrap":[61],"the":[62,74,138,148,203,208],"self-evolution":[63,81],"process.":[64],"In":[65],"this":[66],"work,":[67],"we":[68],"present":[69],"Multi-model":[70],"Multimodal":[71],"Zero":[72],"(MM-Zero),":[73],"first":[75],"RL-based":[76],"framework":[77,99],"achieve":[79],"zero-data":[80],"VLM":[83,180],"reasoning.":[84],"Moving":[85],"beyond":[86,207],"prior":[87],"dual-role":[88],"(Proposer":[89],"Solver)":[91],"setups,":[92],"MM-Zero":[93,178,190],"introduces":[94],"multi-role":[96],"self-evolving":[97,196],"training":[98],"comprising":[100],"three":[101,143],"specialized":[102],"roles:":[103],"Proposer":[105],"generates":[107],"abstract":[108],"concepts":[110,119],"formulates":[112],"questions;":[113],"Coder":[115],"translates":[117],"these":[118],"into":[120],"executable":[121],"code":[122],"(e.g.,":[123],"Python,":[124],"SVG)":[125],"render":[127],"images;":[129],"Solver":[132],"performs":[134],"multimodal":[135,188,200],"reasoning":[136,181],"over":[137],"generated":[139],"content.":[141],"All":[142],"roles":[144],"are":[145],"initialized":[146],"same":[149],"base":[150],"model":[151],"trained":[153],"using":[154],"Group":[155],"Relative":[156],"Policy":[157],"Optimization":[158],"(GRPO),":[159],"carefully":[161],"designed":[162],"reward":[163],"mechanisms":[164],"integrate":[166],"execution":[167],"feedback,":[168],"verification,":[170],"difficulty":[172],"balancing.":[173],"Our":[174],"experiments":[175],"show":[176],"improves":[179],"performance":[182],"across":[183],"wide":[185],"range":[186],"of":[187,205],"benchmarks.":[189],"establishes":[191],"scalable":[193],"path":[194],"toward":[195],"multi-model":[197],"systems":[198],"models,":[201],"extending":[202],"frontier":[204],"self-improvement":[206],"conventional":[209],"two-model":[210],"paradigm.":[211]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-03-12T00:00:00"}