{"id":"https://openalex.org/W7165671954","doi":"https://doi.org/10.48550/arxiv.2606.21775","title":"Beyond the Next Step: Variable-Length Latent World Models for Long-Horizon Planning","display_name":"Beyond the Next Step: Variable-Length Latent World Models for Long-Horizon Planning","publication_year":2026,"publication_date":"2026-06-19","ids":{"openalex":"https://openalex.org/W7165671954","doi":"https://doi.org/10.48550/arxiv.2606.21775"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.21775","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.21775","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2606.21775","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5017499477","display_name":"Tianqi Du","orcid":"https://orcid.org/0000-0002-1894-9607"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Du, Tianqi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036390522","display_name":"Q. Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Qi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139190224","display_name":"Yifei Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yifei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139160417","display_name":"Yisen Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yisen","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":0.2953999936580658,"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":0.2953999936580658,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.15649999678134918,"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/T10906","display_name":"AI-based Problem Solving and Planning","score":0.10989999771118164,"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/action","display_name":"Action (physics)","score":0.5763000249862671},{"id":"https://openalex.org/keywords/latent-variable","display_name":"Latent variable","score":0.5680999755859375},{"id":"https://openalex.org/keywords/exploit","display_name":"Exploit","score":0.5072000026702881},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.4657999873161316},{"id":"https://openalex.org/keywords/space","display_name":"Space (punctuation)","score":0.41609999537467957},{"id":"https://openalex.org/keywords/control","display_name":"Control (management)","score":0.3476000130176544},{"id":"https://openalex.org/keywords/simple","display_name":"Simple (philosophy)","score":0.34529998898506165}],"concepts":[{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.6766999959945679},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6294999718666077},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5985000133514404},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.5763000249862671},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.5680999755859375},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.5072000026702881},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.4657999873161316},{"id":"https://openalex.org/C2778572836","wikidata":"https://www.wikidata.org/wiki/Q380933","display_name":"Space (punctuation)","level":2,"score":0.41609999537467957},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.3476000130176544},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.34529998898506165},{"id":"https://openalex.org/C65965080","wikidata":"https://www.wikidata.org/wiki/Q1806885","display_name":"Latent variable model","level":3,"score":0.34380000829696655},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.32409998774528503},{"id":"https://openalex.org/C59404180","wikidata":"https://www.wikidata.org/wiki/Q17013334","display_name":"Feature learning","level":2,"score":0.30469998717308044},{"id":"https://openalex.org/C182365436","wikidata":"https://www.wikidata.org/wiki/Q50701","display_name":"Variable (mathematics)","level":2,"score":0.2720000147819519},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.26919999718666077},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.26190000772476196},{"id":"https://openalex.org/C45804977","wikidata":"https://www.wikidata.org/wiki/Q7239673","display_name":"Predictive modelling","level":2,"score":0.2517000138759613}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.21775","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.21775","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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.2606.21775","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.21775","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":null,"issn":null,"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/11","display_name":"Sustainable cities and communities","score":0.6230396032333374}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Recently,":[0],"world":[1,31,52,184,225],"models":[2,16,32,53],"have":[3],"emerged":[4],"as":[5],"a":[6,74,94,137,212],"promising":[7],"paradigm":[8,216],"for":[9,65,217],"building":[10],"intelligent":[11],"agents":[12],"by":[13,41],"learning":[14,42],"predictive":[15,170],"that":[17,96,141,178,209],"estimate":[18],"future":[19,100],"environment":[20],"states":[21,102],"conditioned":[22,103],"on":[23,56,104,114,173,201],"observations":[24],"and":[25,59,73,79,221],"actions.":[26],"In":[27],"particular,":[28],"JEPA-style":[29],"latent":[30,51,101,182,224],"provide":[33],"an":[34],"efficient":[35],"alternative":[36],"to":[37,70,98,127,152,163,165],"pixel":[38],"space":[39,183],"prediction":[40,58,220],"action-conditioned":[43],"dynamics":[44,151],"in":[45,223],"compact":[46],"representation":[47],"spaces.":[48],"However,":[49],"existing":[50],"typically":[54],"rely":[55],"one-step":[57,115],"must":[60],"be":[61],"recursively":[62],"rolled":[63],"out":[64],"long-horizon":[66,174,219],"planning,":[67],"which":[68],"leads":[69],"compounding":[71],"errors":[72],"mismatch":[75],"between":[76],"training":[77,112,139],"objectives":[78],"downstream":[80],"planning":[81,160,222],"tasks.":[82],"To":[83],"address":[84],"this":[85],"limitation,":[86],"we":[87,158],"propose":[88],"Variable-length":[89],"Latent":[90],"World":[91],"Models":[92],"(VLWMs),":[93],"framework":[95],"learns":[97],"predict":[99],"action":[105,129,145],"sequences":[106],"of":[107,111],"variable":[108],"lengths.":[109],"Instead":[110],"only":[113],"transitions,":[116],"VLWMs":[117,164,179],"directly":[118],"model":[119],"temporally":[120],"extended":[121,204],"dynamics,":[122],"allowing":[123],"the":[124,144,191],"same":[125],"predictor":[126],"evaluate":[128],"plans":[130],"over":[131,190],"different":[132,195],"horizons.":[133],"We":[134],"further":[135],"introduce":[136],"curriculum":[138],"strategy":[140],"progressively":[142],"expands":[143],"horizon,":[146],"stabilizing":[147],"optimization":[148],"from":[149],"short-range":[150],"long-range":[153],"prediction.":[154],"At":[155],"test":[156],"time,":[157],"design":[159],"methods":[161],"tailored":[162],"better":[166],"exploit":[167],"their":[168],"variable-length":[169],"capabilities.":[171],"Experiments":[172],"control":[175],"tasks":[176,202],"show":[177],"significantly":[180],"improve":[181],"models,":[185],"achieving":[186],"13\\%":[187],"average":[188],"improvement":[189],"state-of-the-art":[192],"LeWM":[193],"across":[194],"datasets,":[196],"with":[197],"especially":[198],"large":[199],"gains":[200],"requiring":[203],"planning.":[205],"These":[206],"results":[207],"suggest":[208],"VLWM":[210],"provides":[211],"simple":[213],"yet":[214],"effective":[215],"improving":[218],"models.":[226]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-24T00:00:00"}