{"id":"https://openalex.org/W7163179828","doi":"https://doi.org/10.48550/arxiv.2606.02372","title":"COMAP: Co-Evolving World Models and Agent Policies for LLM Agents","display_name":"COMAP: Co-Evolving World Models and Agent Policies for LLM Agents","publication_year":2026,"publication_date":"2026-06-01","ids":{"openalex":"https://openalex.org/W7163179828","doi":"https://doi.org/10.48550/arxiv.2606.02372"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.02372","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.02372","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":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.2606.02372","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5123084769","display_name":"Youwei Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Youwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137680804","display_name":"Jian Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137626135","display_name":"Hanlin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Hanlin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5137678169","display_name":"Wenjie Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Wenjie","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":0.3984000086784363,"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":0.3984000086784363,"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":0.13539999723434448,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.043299999088048935,"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/task","display_name":"Task (project management)","score":0.6014000177383423},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5083000063896179},{"id":"https://openalex.org/keywords/reliability","display_name":"Reliability (semiconductor)","score":0.48899999260902405},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.47350001335144043},{"id":"https://openalex.org/keywords/state","display_name":"State (computer science)","score":0.41370001435279846},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.3944999873638153},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.3880000114440918}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6833999752998352},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6014000177383423},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5083000063896179},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.48899999260902405},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.47350001335144043},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4693000018596649},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.41370001435279846},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.3944999873638153},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.39329999685287476},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.3880000114440918},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.3481999933719635},{"id":"https://openalex.org/C65682993","wikidata":"https://www.wikidata.org/wiki/Q1056451","display_name":"Reflection (computer programming)","level":2,"score":0.33970001339912415},{"id":"https://openalex.org/C41550386","wikidata":"https://www.wikidata.org/wiki/Q529909","display_name":"Multi-agent system","level":2,"score":0.28839999437332153},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.273499995470047},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.260699987411499},{"id":"https://openalex.org/C136197465","wikidata":"https://www.wikidata.org/wiki/Q1729295","display_name":"Variety (cybernetics)","level":2,"score":0.25949999690055847},{"id":"https://openalex.org/C2778348673","wikidata":"https://www.wikidata.org/wiki/Q739302","display_name":"Production (economics)","level":2,"score":0.25769999623298645},{"id":"https://openalex.org/C3020493868","wikidata":"https://www.wikidata.org/wiki/Q55631277","display_name":"Real world data","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.02372","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.02372","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":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.2606.02372","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.02372","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.7461325526237488}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Equipping":[0],"language":[1],"agents":[2],"with":[3,153],"world":[4,21,71,84,121,164],"models":[5,22,72],"enables":[6],"them":[7,29],"to":[8,32,118,127,172],"anticipate":[9],"environment":[10],"dynamics":[11],"and":[12,73,93,106,141,170],"evaluate":[13],"candidate":[14,91],"actions":[15],"before":[16],"execution.":[17],"However,":[18],"existing":[19],"textual":[20,70],"are":[23,115],"typically":[24],"fixed":[25],"after":[26],"training,":[27],"preventing":[28],"from":[30],"adapting":[31],"the":[33,83,94,101,120,130,159,163],"on-policy":[34,113],"state-action":[35],"distributions":[36],"induced":[37],"by":[38,99],"an":[39],"evolving":[40,132],"agent.":[41],"Meanwhile,":[42],"agent-improvement":[43],"methods":[44],"often":[45],"rely":[46],"on":[47],"external":[48],"rewards":[49],"or":[50],"verifiers,":[51],"limiting":[52],"their":[53],"applicability":[54],"in":[55],"realistic":[56],"interactive":[57],"environments.":[58],"In":[59],"this":[60,104],"paper,":[61],"we":[62],"propose":[63],"COMAP,":[64],"a":[65],"novel":[66],"framework":[67],"that":[68,158],"co-evolves":[69],"agent":[74,95],"policies":[75],"through":[76],"closed-loop":[77],"interaction.":[78],"At":[79],"each":[80],"decision":[81],"step,":[82],"model":[85,122],"predicts":[86],"future":[87],"state":[88],"feedback":[89,105],"for":[90],"actions,":[92],"performs":[96],"future-aware":[97],"reflection":[98],"estimating":[100],"reliability":[102],"of":[103],"refining":[107],"its":[108],"action":[109],"accordingly.":[110],"The":[111],"resulting":[112],"trajectories":[114],"then":[116],"used":[117],"update":[119],"via":[123],"self-distillation,":[124],"allowing":[125],"it":[126],"better":[128],"match":[129],"agent's":[131],"interaction":[133],"distribution.":[134],"Across":[135],"embodied":[136],"task":[137],"planning,":[138],"Web":[139],"navigation,":[140],"tool-use":[142],"benchmarks,":[143],"COMAP":[144],"consistently":[145],"outperforms":[146],"competitive":[147],"baselines,":[148],"e.g.,":[149],"+16.75%":[150],"relative":[151],"improvement":[152],"Qwen3-4B.":[154],"Further":[155],"analyses":[156],"show":[157],"co-evolutionary":[160],"loop":[161],"improves":[162],"model's":[165],"prediction":[166],"accuracy":[167],"over":[168],"time":[169],"leads":[171],"more":[173],"effective":[174],"long-horizon":[175],"decision-making.":[176],"Our":[177],"code":[178],"is":[179],"available":[180],"at:":[181],"https://github.com/loyiv/CoMAP.":[182]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-03T00:00:00"}