{"id":"https://openalex.org/W7161235788","doi":"https://doi.org/10.48550/arxiv.2605.14212","title":"MetaAgent-X : Breaking the Ceiling of Automatic Multi-Agent Systems via End-to-End Reinforcement Learning","display_name":"MetaAgent-X : Breaking the Ceiling of Automatic Multi-Agent Systems via End-to-End Reinforcement Learning","publication_year":2026,"publication_date":"2026-05-14","ids":{"openalex":"https://openalex.org/W7161235788","doi":"https://doi.org/10.48550/arxiv.2605.14212"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.14212","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14212","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.2605.14212","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5124894589","display_name":"Yaolun Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Yaolun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136237088","display_name":"Yujie Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Yujie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136244927","display_name":"Nan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Nan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007402964","display_name":"Yiran Wu","orcid":"https://orcid.org/0000-0001-6790-7473"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yiran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136254748","display_name":"Jiayu Chang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chang, Jiayu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136247059","display_name":"Yizhao Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Yizhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136214819","display_name":"Qingyun Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Qingyun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136238245","display_name":"Jishen Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Jishen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136224316","display_name":"Huazheng Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Huazheng","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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.21850000321865082,"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"}},"topics":[{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.21850000321865082,"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/T12784","display_name":"Modular Robots and Swarm Intelligence","score":0.17479999363422394,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10639","display_name":"Advanced Software Engineering Methodologies","score":0.14589999616146088,"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/executor","display_name":"Executor","score":0.982699990272522},{"id":"https://openalex.org/keywords/reinforcement-learning","display_name":"Reinforcement learning","score":0.7820000052452087},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.6919999718666077},{"id":"https://openalex.org/keywords/workflow","display_name":"Workflow","score":0.5738000273704529},{"id":"https://openalex.org/keywords/stability","display_name":"Stability (learning theory)","score":0.47620001435279846},{"id":"https://openalex.org/keywords/ceiling","display_name":"Ceiling (cloud)","score":0.4480000138282776},{"id":"https://openalex.org/keywords/bridging","display_name":"Bridging (networking)","score":0.4309999942779541},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.37700000405311584}],"concepts":[{"id":"https://openalex.org/C180591056","wikidata":"https://www.wikidata.org/wiki/Q654437","display_name":"Executor","level":2,"score":0.982699990272522},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7820000052452087},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7809000015258789},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.6919999718666077},{"id":"https://openalex.org/C177212765","wikidata":"https://www.wikidata.org/wiki/Q627335","display_name":"Workflow","level":2,"score":0.5738000273704529},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4799000024795532},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.47620001435279846},{"id":"https://openalex.org/C2777489069","wikidata":"https://www.wikidata.org/wiki/Q1589822","display_name":"Ceiling (cloud)","level":2,"score":0.4480000138282776},{"id":"https://openalex.org/C174348530","wikidata":"https://www.wikidata.org/wiki/Q188635","display_name":"Bridging (networking)","level":2,"score":0.4309999942779541},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.37700000405311584},{"id":"https://openalex.org/C3019831412","wikidata":"https://www.wikidata.org/wiki/Q5778278","display_name":"Fully automatic","level":2,"score":0.35260000824928284},{"id":"https://openalex.org/C115901376","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automation","level":2,"score":0.34940001368522644},{"id":"https://openalex.org/C167123822","wikidata":"https://www.wikidata.org/wiki/Q184199","display_name":"Automatic control","level":2,"score":0.30250000953674316},{"id":"https://openalex.org/C197129107","wikidata":"https://www.wikidata.org/wiki/Q1921621","display_name":"Merge (version control)","level":2,"score":0.2964000105857849},{"id":"https://openalex.org/C117623542","wikidata":"https://www.wikidata.org/wiki/Q621974","display_name":"Automatic target recognition","level":3,"score":0.289000004529953},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.28360000252723694},{"id":"https://openalex.org/C133731056","wikidata":"https://www.wikidata.org/wiki/Q4917288","display_name":"Control engineering","level":1,"score":0.2777999937534332},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.27649998664855957},{"id":"https://openalex.org/C4924752","wikidata":"https://www.wikidata.org/wiki/Q184148","display_name":"Plug-in","level":2,"score":0.27230000495910645},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.27149999141693115},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.26019999384880066},{"id":"https://openalex.org/C2777211547","wikidata":"https://www.wikidata.org/wiki/Q17141490","display_name":"Training (meteorology)","level":2,"score":0.2547000050544739},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.25049999356269836}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.14212","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14212","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.2605.14212","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.14212","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":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Automatic":[0],"multi-agent":[1],"systems":[2],"aim":[3],"to":[4,110,130],"instantiate":[5],"agent":[6],"workflows":[7],"without":[8],"relying":[9],"on":[10],"manually":[11],"designed":[12],"or":[13,31],"fixed":[14],"orchestration.":[15],"However,":[16],"existing":[17,124],"automatic":[18,73,125,147,160],"MAS":[19,74,81,126,148,161],"approaches":[20],"remain":[21],"only":[22],"partially":[23],"adaptive:":[24],"they":[25],"either":[26],"perform":[27],"training-free":[28],"test-time":[29],"search":[30],"optimize":[32],"the":[33,49,116],"meta-level":[34],"designer":[35,91,138],"while":[36],"keeping":[37],"downstream":[38],"execution":[39,83],"agents":[40],"frozen,":[41],"which":[42],"creating":[43],"a":[44,151,163],"frozen-executor":[45],"ceiling":[46],"and":[47,54,76,86,92,98,107,114,139,144,169],"leaving":[48],"end-to-end":[50,66,158],"training":[51,112],"of":[52,118],"self-designing":[53,168],"self-executing":[55,170],"agentic":[56,171],"models":[57],"unexplored.":[58],"To":[59,95],"address":[60],"this,":[61],"we":[62,101],"introduce":[63],"MetaAgent-X,":[64],"an":[65],"reinforcement":[67],"learning":[68,149],"framework":[69],"that":[70,136,145],"jointly":[71],"optimizes":[72],"design":[75],"execution.":[77],"MetaAgent-X":[78,121],"enables":[79],"script-based":[80],"generation,":[82],"rollout":[84],"collection,":[85],"credit":[87],"assignment":[88],"for":[89,166],"both":[90,137],"executor":[93,140],"trajectories.":[94],"support":[96],"stable":[97],"scalable":[99],"optimization,":[100],"propose":[102],"Executor":[103],"Designer":[104],"Hierarchical":[105],"Rollout":[106],"Stagewise":[108],"Co-evolution":[109],"improve":[111,141],"stability":[113],"expose":[115],"dynamics":[117],"designer-executor":[119],"co-evolution.":[120],"consistently":[122],"outperforms":[123],"baselines,":[127],"achieving":[128],"up":[129],"21.7%":[131],"gains.":[132],"Comprehensive":[133],"ablations":[134],"show":[135],"throughout":[142],"training,":[143],"effective":[146],"follows":[150],"stagewise":[152],"co-evolution":[153],"process.":[154],"These":[155],"results":[156],"establish":[157],"trainable":[159],"as":[162],"practical":[164],"paradigm":[165],"building":[167],"models.":[172]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-16T00:00:00"}