{"id":"https://openalex.org/W7164911105","doi":"https://doi.org/10.48550/arxiv.2606.15160","title":"DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning","display_name":"DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning","publication_year":2026,"publication_date":"2026-06-13","ids":{"openalex":"https://openalex.org/W7164911105","doi":"https://doi.org/10.48550/arxiv.2606.15160"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.15160","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.15160","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":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.2606.15160","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5138735402","display_name":"David Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, David","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5010380116","display_name":"Lianlei Shan","orcid":"https://orcid.org/0000-0002-4648-8246"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Shan, Lianlei","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":0.9513000249862671,"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":0.9513000249862671,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.014000000432133675,"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/T10028","display_name":"Topic Modeling","score":0.005900000222027302,"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/visual-reasoning","display_name":"Visual reasoning","score":0.5382000207901001},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.4449999928474426},{"id":"https://openalex.org/keywords/set","display_name":"Set (abstract data type)","score":0.423799991607666},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.4083999991416931},{"id":"https://openalex.org/keywords/reasoning-system","display_name":"Reasoning system","score":0.382999986410141},{"id":"https://openalex.org/keywords/memory-footprint","display_name":"Memory footprint","score":0.3691999912261963},{"id":"https://openalex.org/keywords/latent-variable","display_name":"Latent variable","score":0.3671000003814697},{"id":"https://openalex.org/keywords/merge","display_name":"Merge (version control)","score":0.36000001430511475}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.724399983882904},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6000999808311462},{"id":"https://openalex.org/C2777508537","wikidata":"https://www.wikidata.org/wiki/Q7936620","display_name":"Visual reasoning","level":2,"score":0.5382000207901001},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4661000072956085},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.4449999928474426},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.423799991607666},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.4083999991416931},{"id":"https://openalex.org/C89288958","wikidata":"https://www.wikidata.org/wiki/Q7301504","display_name":"Reasoning system","level":2,"score":0.382999986410141},{"id":"https://openalex.org/C74912251","wikidata":"https://www.wikidata.org/wiki/Q6815727","display_name":"Memory footprint","level":2,"score":0.3691999912261963},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.3671000003814697},{"id":"https://openalex.org/C197129107","wikidata":"https://www.wikidata.org/wiki/Q1921621","display_name":"Merge (version control)","level":2,"score":0.36000001430511475},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.3580999970436096},{"id":"https://openalex.org/C527412718","wikidata":"https://www.wikidata.org/wiki/Q855395","display_name":"Interpretation (philosophy)","level":2,"score":0.3549000024795532},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.3495999872684479},{"id":"https://openalex.org/C170133592","wikidata":"https://www.wikidata.org/wiki/Q1806883","display_name":"Latent semantic analysis","level":2,"score":0.33250001072883606},{"id":"https://openalex.org/C107848011","wikidata":"https://www.wikidata.org/wiki/Q4680756","display_name":"Adaptive reasoning","level":4,"score":0.29809999465942383},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.29350000619888306},{"id":"https://openalex.org/C97364631","wikidata":"https://www.wikidata.org/wiki/Q484284","display_name":"Deductive reasoning","level":2,"score":0.2791999876499176},{"id":"https://openalex.org/C86827895","wikidata":"https://www.wikidata.org/wiki/Q7098582","display_name":"Opportunistic reasoning","level":4,"score":0.27639999985694885},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.26109999418258667},{"id":"https://openalex.org/C103057564","wikidata":"https://www.wikidata.org/wiki/Q4751139","display_name":"Analytic reasoning","level":3,"score":0.2605000138282776}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.15160","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.15160","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":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.2606.15160","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.15160","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":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":{"Reasoning":[0],"capabilities":[1],"of":[2,124,139],"multimodal":[3,57,109,207],"large":[4],"language":[5],"models":[6],"(MLLMs)":[7],"have":[8],"improved":[9],"considerably":[10],"in":[11,129,219],"recent":[12],"years.":[13],"Existing":[14],"approaches":[15],"typically":[16],"rely":[17],"on":[18,148,205],"explicit":[19],"chain-of-thought":[20],"or":[21,49,69,191],"continuous":[22,130],"latent-space":[23,114],"trajectories":[24],"to":[25,64,74,88,188],"enhance":[26],"multi-step":[27],"reasoning.":[28],"However,":[29],"these":[30],"methods":[31,215],"generally":[32],"assume":[33],"that":[34,112,179,211],"an":[35],"input":[36],"admits":[37],"a":[38,46,51,108,122,135,167,174],"single":[39],"latent":[40,126,131,145,163],"interpretation":[41,138],"and":[42,98,143,172,200],"unfold":[43,144],"reasoning":[44,80,110,115,146,164,193,208],"along":[45],"fixed":[47],"path":[48],"under":[50],"uniform":[52,79],"computation":[53,182],"budget.":[54],"In":[55],"real-world":[56],"settings,":[58],"visual":[59,141],"observations":[60],"are":[61],"often":[62],"subject":[63],"occlusion,":[65],"blur,":[66],"viewpoint":[67],"variation,":[68],"semantic":[70],"ambiguity,":[71],"giving":[72],"rise":[73],"multiple":[75,90,206],"plausible":[76,137],"interpretations.":[77],"A":[78],"strategy":[81],"not":[82],"only":[83],"limits":[84],"the":[85,140,162],"model's":[86],"ability":[87],"explore":[89],"hypotheses":[91,128],"but":[92],"also":[93],"incurs":[94],"high":[95],"memory":[96,198,223],"usage":[97,224],"rollout":[99,202],"cost.":[100],"We":[101],"present":[102],"DLWM":[103,212],"(Diverse":[104],"Latent":[105],"World":[106],"Models),":[107],"framework":[111],"combines":[113],"with":[116],"reinforcement":[117,176],"learning.":[118],"First,":[119],"we":[120,160],"construct":[121],"set":[123],"diverse":[125],"world":[127],"space,":[132],"each":[133,149],"capturing":[134],"different":[136],"input,":[142],"independently":[147],"hypothesis.":[150],"An":[151],"orthogonality-based":[152],"diversity":[153],"regularizer":[154],"explicitly":[155],"prevents":[156],"hypothesis":[157],"collapse.":[158],"Second,":[159],"formulate":[161],"process":[165],"as":[166],"resource-constrained":[168],"sequential":[169],"decision":[170],"problem":[171],"introduce":[173],"resource-aware":[175],"learning":[177],"policy":[178],"adaptively":[180],"allocates":[181],"across":[183],"hypotheses,":[184],"dynamically":[185],"deciding":[186],"whether":[187],"expand,":[189],"terminate,":[190],"merge":[192],"paths,":[194],"thereby":[195],"substantially":[196],"reducing":[197,222],"footprint":[199],"improving":[201],"efficiency.":[203],"Experiments":[204],"benchmarks":[209],"demonstrate":[210],"outperforms":[213],"existing":[214],"by":[216,225],"2-5":[217],"points":[218],"accuracy":[220],"while":[221],"24%.":[226]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-17T00:00:00"}