{"id":"https://openalex.org/W7215108157","doi":"https://doi.org/10.48550/arxiv.2609.40362","title":"Multimodal Flow: Unified Flow Modeling of Language and Vision in Embedding Spaces","display_name":"Multimodal Flow: Unified Flow Modeling of Language and Vision in Embedding Spaces","publication_year":2026,"publication_date":"2026-09-30","ids":{"openalex":"https://openalex.org/W7215108157","doi":"https://doi.org/10.48550/arxiv.2609.40362"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.40362","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.40362","pdf_url":"https://arxiv.org/pdf/2609.40362","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.40362","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5065454973","display_name":"Hongyuan Tao","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Tao, Hongyuan","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5154170946","display_name":"Xinggang Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wang, Xinggang","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153984755","display_name":"Lianghui Zhu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhu, Lianghui","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5154151850","display_name":"Yongkang Li","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li, Yongkang","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5154253225","display_name":"Yunchao Wei","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I21193070","display_name":"Beijing Jiaotong University","ror":"https://ror.org/01yj56c84","country_code":"CN","type":"education","lineage":["https://openalex.org/I21193070"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wei, Yunchao","raw_affiliation_strings":["Beijing Jiaotong University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Beijing Jiaotong University","institution_ids":["https://openalex.org/I21193070"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153875997","display_name":"Bin Feng","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I47720641","display_name":"Huazhong University of Science and Technology","ror":"https://ror.org/00p991c53","country_code":"CN","type":"education","lineage":["https://openalex.org/I47720641"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Feng, Bin","raw_affiliation_strings":["Huazhong University of Science and Technology"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Huazhong University of Science and Technology","institution_ids":["https://openalex.org/I47720641"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5154208535","display_name":"Shaoyu Chen","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4401726824","display_name":"Horizon Robotics (China)","ror":"https://ror.org/05cmv6g68","country_code":"CN","type":"company","lineage":["https://openalex.org/I4401726824"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen, Shaoyu","raw_affiliation_strings":["Horizon Robotics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Horizon Robotics","institution_ids":["https://openalex.org/I4401726824"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5153806309","display_name":"Qian Zhang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4401726824","display_name":"Horizon Robotics (China)","ror":"https://ror.org/05cmv6g68","country_code":"CN","type":"company","lineage":["https://openalex.org/I4401726824"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhang, Qian","raw_affiliation_strings":["Horizon Robotics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Horizon Robotics","institution_ids":["https://openalex.org/I4401726824"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5154037221","display_name":"Chang Huang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4401726824","display_name":"Horizon Robotics (China)","ror":"https://ror.org/05cmv6g68","country_code":"CN","type":"company","lineage":["https://openalex.org/I4401726824"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huang, Chang","raw_affiliation_strings":["Horizon Robotics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Horizon Robotics","institution_ids":["https://openalex.org/I4401726824"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5153962276","display_name":"Kai Yu","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I4401726824","display_name":"Horizon Robotics (China)","ror":"https://ror.org/05cmv6g68","country_code":"CN","type":"company","lineage":["https://openalex.org/I4401726824"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yu, Kai","raw_affiliation_strings":["Horizon Robotics"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Horizon Robotics","institution_ids":["https://openalex.org/I4401726824"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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":1.0,"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":1.0,"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":0.0,"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":0.0,"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/multimodal-pretraining","display_name":"multimodal pre-training","score":0.6460000276565552},{"id":"https://openalex.org/keywords/flow-matching","display_name":"flow matching","score":0.8700000047683716},{"id":"https://openalex.org/keywords/cross-modal-attention","display_name":"cross-modal attention","score":0.777999997138977},{"id":"https://openalex.org/keywords/image-text-generation","display_name":"image-text generation","score":0.3400000035762787}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7674999833106995},{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.7110000252723694},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5496000051498413},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5098000168800354},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.45089998841285706},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.4345000088214874},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.42719998955726624},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.42570000886917114},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.4083000123500824},{"id":"https://openalex.org/C45493050","wikidata":"https://www.wikidata.org/wiki/Q7884934","display_name":"Unified Model","level":2,"score":0.3736000061035156},{"id":"https://openalex.org/C9652623","wikidata":"https://www.wikidata.org/wiki/Q190109","display_name":"Field (mathematics)","level":2,"score":0.3686999976634979},{"id":"https://openalex.org/C155542232","wikidata":"https://www.wikidata.org/wiki/Q736111","display_name":"Optical flow","level":3,"score":0.3675000071525574},{"id":"https://openalex.org/C38349280","wikidata":"https://www.wikidata.org/wiki/Q1434290","display_name":"Flow (mathematics)","level":2,"score":0.33719998598098755},{"id":"https://openalex.org/C2985264121","wikidata":"https://www.wikidata.org/wiki/Q216320","display_name":"Continuous flow","level":2,"score":0.32910001277923584},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.32339999079704285},{"id":"https://openalex.org/C18555067","wikidata":"https://www.wikidata.org/wiki/Q8375051","display_name":"Joint (building)","level":2,"score":0.3163999915122986},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.2939000129699707},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2924000024795532},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.28529998660087585},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.26899999380111694},{"id":"https://openalex.org/C135641252","wikidata":"https://www.wikidata.org/wiki/Q738567","display_name":"Multimodal interaction","level":2,"score":0.26409998536109924},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.260699987411499},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.2574000060558319}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.40362","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.40362","pdf_url":"https://arxiv.org/pdf/2609.40362","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.40362","doi":"https://doi.org/10.48550/arxiv.2609.40362","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.40362","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.40362","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.40362","pdf_url":"https://arxiv.org/pdf/2609.40362","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7215108157.pdf","grobid_xml":"https://content.openalex.org/works/W7215108157.grobid-xml"},"referenced_works_count":30,"referenced_works":["https://openalex.org/W4288099666","https://openalex.org/W4303647933","https://openalex.org/W4306802991","https://openalex.org/W4366208220","https://openalex.org/W4386185600","https://openalex.org/W4391631722","https://openalex.org/W4392678285","https://openalex.org/W4396945138","https://openalex.org/W4397049097","https://openalex.org/W4403799526","https://openalex.org/W4403853618","https://openalex.org/W4406975803","https://openalex.org/W4407815020","https://openalex.org/W4414940327","https://openalex.org/W4414969226","https://openalex.org/W4415061330","https://openalex.org/W4416982486","https://openalex.org/W4417082432","https://openalex.org/W4417461713","https://openalex.org/W6929300824","https://openalex.org/W6948058629","https://openalex.org/W6948254895","https://openalex.org/W6967290142","https://openalex.org/W7083819494","https://openalex.org/W7114776436","https://openalex.org/W7125500369","https://openalex.org/W7160596537","https://openalex.org/W7160895307","https://openalex.org/W7162788221","https://openalex.org/W7202417173"],"related_works":["https://openalex.org/W7223242516","https://openalex.org/W7170145606","https://openalex.org/W7160831703","https://openalex.org/W7119040870","https://openalex.org/W7103136884","https://openalex.org/W4415312711","https://openalex.org/W4414742964","https://openalex.org/W7133498487","https://openalex.org/W7128779974","https://openalex.org/W7196960623"],"abstract_inverted_index":{"We":[0,146],"present":[1],"Multimodal":[2,69,207],"Flow,":[3],"a":[4,39,59,72,82,108,225],"fully":[5,227],"continuous":[6,33,52,74,79,95,219,228],"generative":[7,61],"model":[8,18,131,238],"of":[9,177],"language":[10,20,30],"and":[11,21,48,57,91,101,140,149,158,181,183,188,204,213,237],"vision.":[12],"Most":[13],"unified":[14,73,193,231],"multimodal":[15,67,78,153,165,232],"models":[16,194],"either":[17],"both":[19],"quantized":[22],"images":[23,92],"as":[24,93,224],"discrete":[25,29,214],"tokens":[26],"or":[27],"combine":[28],"prediction":[31],"with":[32,81,192],"image":[34],"generation.":[35],"The":[36,43,105,130,234],"former":[37],"introduces":[38,71],"visual":[40,102],"quantization":[41],"bottleneck.":[42],"latter":[44],"requires":[45],"modality-dependent":[46],"objectives":[47],"sampling":[49],"procedures.":[50],"Fully":[51],"modeling":[53,223],"avoids":[54],"these":[55,113],"trade-offs":[56],"enables":[58,120],"shared":[60,83],"process,":[62],"but":[63],"remains":[64],"underexplored":[65],"for":[66,230],"pretraining.":[68],"Flow":[70,116,208],"architecture":[75],"that":[76],"integrates":[77],"representations":[80],"chunk-causal":[84],"flow":[85,222],"backbone.":[86],"It":[87],"organizes":[88],"text":[89],"blocks":[90],"ordered":[94],"hyperchunks,":[96],"preserving":[97],"textual":[98],"token":[99],"order":[100],"spatial":[103],"structure.":[104],"backbone":[106],"learns":[107],"single":[109],"vector":[110],"field":[111],"over":[112],"hyperchunks":[114,142],"through":[115],"Matching.":[117],"Joint":[118],"attention":[119],"cross-modal":[121],"interaction,":[122],"while":[123],"modality-specific":[124],"feed-forward":[125],"networks":[126],"process":[127],"each":[128],"modality.":[129],"predicts":[132],"multiple":[133],"target":[134],"chunks":[135],"in":[136],"parallel":[137],"during":[138],"training":[139],"generates":[141],"sequentially":[143],"at":[144,242],"inference.":[145],"instantiate":[147],"MF-1":[148,172],"pretrain":[150],"it":[151],"on":[152,196],"data.":[154,199],"Across":[155],"0.6B,":[156],"1.2B,":[157],"1.6B":[159],"scales,":[160],"continued":[161],"pretraining":[162,170],"consistently":[163],"improves":[164],"modeling.":[166,233],"With":[167],"only":[168],"150B":[169],"tokens,":[171],"achieves":[173],"an":[174],"average":[175],"score":[176],"82.8":[178],"across":[179,185],"GenEval":[180],"DPG-Bench":[182],"75.3":[184],"VQAv2,":[186],"MMBench,":[187],"POPE,":[189],"remaining":[190],"competitive":[191],"trained":[195],"substantially":[197],"more":[198],"Under":[200],"matched":[201],"data,":[202],"optimization,":[203],"parameter":[205],"budgets,":[206],"further":[209],"outperforms":[210],"representative":[211],"hybrid":[212],"models.":[215],"These":[216],"results":[217],"establish":[218],"chunk-based":[220],"embedding":[221],"new":[226],"paradigm":[229],"related":[235],"code":[236],"are":[239],"publicly":[240],"released":[241],"https://github.com/hustvl/Multimodal-Flow.":[243]},"counts_by_year":[],"updated_date":"2026-11-07T11:25:07.637180","created_date":"2026-10-02T00:00:00"}