{"id":"https://openalex.org/W7128399547","doi":"https://doi.org/10.48550/arxiv.2602.06056","title":"Analyzing Diffusion and Autoregressive Vision Language Models in Multimodal Embedding Space","display_name":"Analyzing Diffusion and Autoregressive Vision Language Models in Multimodal Embedding Space","publication_year":2026,"publication_date":"2026-01-19","ids":{"openalex":"https://openalex.org/W7128399547","doi":"https://doi.org/10.48550/arxiv.2602.06056"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2602.06056","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.06056","pdf_url":"https://arxiv.org/pdf/2602.06056","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","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2602.06056","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122276457","display_name":"Zihang Wang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zihang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123631718","display_name":"Siyue Zhang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Siyue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123677054","display_name":"Yilun Zhao","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Yilun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5124759277","display_name":"Jingyi Yang","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Jingyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123657836","display_name":"Tingyu Song","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Tingyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5123004532","display_name":"Anh Tuan Luu","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luu, Anh Tuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5125402111","display_name":"Chen Zhao","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Chen","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":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/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"}},{"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"}}],"keywords":[{"id":"https://openalex.org/keywords/diffusion-language-models","display_name":"diffusion language models","score":0.8090000152587891},{"id":"https://openalex.org/keywords/image-text-alignment","display_name":"image-text alignment","score":0.8270000219345093},{"id":"https://openalex.org/keywords/visual-question-answering","display_name":"visual question answering","score":0.828000009059906},{"id":"https://openalex.org/keywords/image-classification","display_name":"image classification","score":0.628000020980835},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"information retrieval","score":0.8889999985694885}],"concepts":[{"id":"https://openalex.org/C41608201","wikidata":"https://www.wikidata.org/wiki/Q980509","display_name":"Embedding","level":2,"score":0.8999999761581421},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.7631999850273132},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6636000275611877},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5468000173568726},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5419999957084656},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4350000023841858},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.36419999599456787},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.3416999876499176},{"id":"https://openalex.org/C2780910867","wikidata":"https://www.wikidata.org/wiki/Q1952416","display_name":"Multimodality","level":2,"score":0.33469998836517334},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.33000001311302185},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.3102000057697296},{"id":"https://openalex.org/C154775046","wikidata":"https://www.wikidata.org/wiki/Q188","display_name":"German","level":2,"score":0.2944999933242798},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.2513999938964844}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2602.06056","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.06056","pdf_url":"https://arxiv.org/pdf/2602.06056","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.2602.06056","doi":"https://doi.org/10.48550/arxiv.2602.06056","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.06056","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":"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2602.06056","doi":null,"is_oa":true,"landing_page_url":"http://arxiv.org/abs/2602.06056","pdf_url":"https://arxiv.org/pdf/2602.06056","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":[],"sustainable_development_goals_aurora":[{"id":"https://metadata.un.org/sdg/4","score":0.7594132423400879,"display_name":"Quality education"}],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4416453495","https://openalex.org/W4403324120","https://openalex.org/W6979241218","https://openalex.org/W4416036849","https://openalex.org/W6967216406","https://openalex.org/W6966793651","https://openalex.org/W7202353063","https://openalex.org/W4392781354","https://openalex.org/W4399198762","https://openalex.org/W4407123378"],"abstract_inverted_index":{"Embedding":[0],"models":[1,22],"are":[2],"a":[3,77],"fundamental":[4],"component":[5],"of":[6,28,100,118],"modern":[7],"AI":[8],"systems":[9],"such":[10,66],"as":[11,58,67,86],"semantic":[12],"search":[13],"and":[14,43,53,70,112,125,157],"retrieval-augmented":[15],"generation.":[16,72],"Recent":[17],"advances":[18],"in":[19,186,194],"large":[20],"foundation":[21],"have":[23,56],"substantially":[24,170],"accelerated":[25],"the":[26,96,164,191],"development":[27],"embedding":[29,89,105,119,196],"models,":[30,63,188],"including":[31],"those":[32],"based":[33],"on":[34,151,155,160],"Large":[35,48],"Language":[36,40,50],"Models":[37,41,51],"(LLMs),":[38],"Vision":[39],"(VLMs),":[42],"Multimodal":[44,54,83,102,110,132],"LLMs.":[45],"More":[46],"recently,":[47],"Diffusion":[49],"(dLLMs)":[52],"dLLMs":[55,84,103,111],"emerged":[57],"competitive":[59],"alternatives":[60],"to":[61],"autoregressive":[62,138],"offering":[64],"advantages":[65],"bidirectional":[68],"attention":[69],"parallel":[71],"This":[73],"progress":[74],"naturally":[75],"raises":[76],"critical":[78],"yet":[79],"unexplored":[80],"question:":[81],"can":[82],"serve":[85],"effective":[87],"multimodal":[88],"models?":[90],"To":[91],"answer":[92],"this,":[93],"we":[94],"present":[95],"first":[97],"systematic":[98],"study":[99],"converting":[101],"into":[104],"models.":[106],"We":[107],"evaluate":[108],"state-of-the-art":[109],"Autoregressive":[113],"VLMs":[114],"across":[115,177],"three":[116],"categories":[117],"tasks:":[120],"classification,":[121,152],"visual":[122],"question":[123],"answering,":[124],"information":[126],"retrieval.":[127],"Our":[128],"results":[129],"show":[130],"that":[131],"dLLM":[133],"embeddings":[134],"generally":[135],"underperform":[136],"their":[137,195],"VLM":[139],"counterparts.":[140],"The":[141],"stronger":[142],"diffusion-based":[143,166,187],"model,":[144,167],"LaViDa,":[145],"lags":[146],"by":[147],"only":[148],"3.5":[149],"points":[150,154,159,176],"2.5":[153],"VQA,":[156],"4.4":[158],"retrieval":[161],"tasks,":[162],"whereas":[163],"other":[165],"MMaDA,":[168],"exhibits":[169],"larger":[171],"performance":[172],"gaps,":[173],"exceeding":[174],"20":[175],"all":[178],"tasks.":[179],"Further":[180],"analysis":[181],"reveals":[182],"insufficient":[183],"image-text":[184],"alignment":[185],"accounting":[189],"for":[190],"observed":[192],"limitations":[193],"performance.":[197]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-02-10T00:00:00"}