{"id":"https://openalex.org/W7163686649","doi":"https://doi.org/10.48550/arxiv.2606.05395","title":"VASO: Formally Verifiable Self-Evolving Skills for Physical AI Agents","display_name":"VASO: Formally Verifiable Self-Evolving Skills for Physical AI Agents","publication_year":2026,"publication_date":"2026-06-03","ids":{"openalex":"https://openalex.org/W7163686649","doi":"https://doi.org/10.48550/arxiv.2606.05395"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.05395","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.05395","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.2606.05395","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5137973501","display_name":"Yunhao Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Yunhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130829178","display_name":"Neel P. Bhatt","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bhatt, Neel P.","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138009933","display_name":"Kevin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Kevin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5018814164","display_name":"Samuel Tetteh","orcid":"https://orcid.org/0000-0002-8989-6346"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tetteh, Samuel","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5137950374","display_name":"Zhangyang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zhangyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138002153","display_name":"Ufuk Topcu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Topcu, Ufuk","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/T10142","display_name":"Formal Methods in Verification","score":0.26809999346733093,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T10142","display_name":"Formal Methods in Verification","score":0.26809999346733093,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T11689","display_name":"Adversarial Robustness in Machine Learning","score":0.1542000025510788,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.14149999618530273,"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/robot","display_name":"Robot","score":0.6222000122070312},{"id":"https://openalex.org/keywords/executable","display_name":"Executable","score":0.5867000222206116},{"id":"https://openalex.org/keywords/interface","display_name":"Interface (matter)","score":0.5374000072479248},{"id":"https://openalex.org/keywords/verifiable-secret-sharing","display_name":"Verifiable secret sharing","score":0.39649999141693115},{"id":"https://openalex.org/keywords/counterexample","display_name":"Counterexample","score":0.3919999897480011},{"id":"https://openalex.org/keywords/model-checking","display_name":"Model checking","score":0.36579999327659607},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.3537999987602234},{"id":"https://openalex.org/keywords/abstraction","display_name":"Abstraction","score":0.3517000079154968}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7170000076293945},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.6222000122070312},{"id":"https://openalex.org/C160145156","wikidata":"https://www.wikidata.org/wiki/Q778586","display_name":"Executable","level":2,"score":0.5867000222206116},{"id":"https://openalex.org/C113843644","wikidata":"https://www.wikidata.org/wiki/Q901882","display_name":"Interface (matter)","level":4,"score":0.5374000072479248},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.48649999499320984},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.45410001277923584},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.43650001287460327},{"id":"https://openalex.org/C85847156","wikidata":"https://www.wikidata.org/wiki/Q59015987","display_name":"Verifiable secret sharing","level":3,"score":0.39649999141693115},{"id":"https://openalex.org/C162838799","wikidata":"https://www.wikidata.org/wiki/Q596077","display_name":"Counterexample","level":2,"score":0.3919999897480011},{"id":"https://openalex.org/C110251889","wikidata":"https://www.wikidata.org/wiki/Q1569697","display_name":"Model checking","level":2,"score":0.36579999327659607},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3537999987602234},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.3517000079154968},{"id":"https://openalex.org/C2779010991","wikidata":"https://www.wikidata.org/wiki/Q2720909","display_name":"Artifact (error)","level":2,"score":0.34540000557899475},{"id":"https://openalex.org/C2776999362","wikidata":"https://www.wikidata.org/wiki/Q2349274","display_name":"Planner","level":2,"score":0.3402000069618225},{"id":"https://openalex.org/C75291252","wikidata":"https://www.wikidata.org/wiki/Q1315756","display_name":"TRACE (psycholinguistics)","level":2,"score":0.3343999981880188},{"id":"https://openalex.org/C111498074","wikidata":"https://www.wikidata.org/wiki/Q173326","display_name":"Formal verification","level":2,"score":0.3343000113964081},{"id":"https://openalex.org/C115903868","wikidata":"https://www.wikidata.org/wiki/Q80993","display_name":"Software engineering","level":1,"score":0.298799991607666},{"id":"https://openalex.org/C2775924081","wikidata":"https://www.wikidata.org/wiki/Q55608371","display_name":"Control (management)","level":2,"score":0.29649999737739563},{"id":"https://openalex.org/C36503486","wikidata":"https://www.wikidata.org/wiki/Q11235244","display_name":"Domain (mathematical analysis)","level":2,"score":0.2948000133037567},{"id":"https://openalex.org/C149810388","wikidata":"https://www.wikidata.org/wiki/Q5374873","display_name":"Emulation","level":2,"score":0.2906999886035919},{"id":"https://openalex.org/C75606506","wikidata":"https://www.wikidata.org/wiki/Q1049183","display_name":"Formal methods","level":2,"score":0.26739999651908875},{"id":"https://openalex.org/C74072328","wikidata":"https://www.wikidata.org/wiki/Q1142726","display_name":"Intelligent agent","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C195344581","wikidata":"https://www.wikidata.org/wiki/Q2555318","display_name":"Automated reasoning","level":2,"score":0.2644999921321869},{"id":"https://openalex.org/C100609095","wikidata":"https://www.wikidata.org/wiki/Q1335050","display_name":"Embodied cognition","level":2,"score":0.25949999690055847}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.05395","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.05395","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.2606.05395","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.05395","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":[{"display_name":"Quality Education","id":"https://metadata.un.org/sdg/4","score":0.7635213732719421}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reusable":[0],"robot":[1,92,114,234],"skills":[2,44,222],"are":[3],"becoming":[4],"the":[5,27,33,149,162,171,209,214],"basic":[6],"units":[7],"through":[8,45],"which":[9],"embodied":[10],"agents":[11],"turn":[12],"open-ended":[13],"instructions":[14],"into":[15,165],"long-horizon":[16],"physical":[17,224],"behavior.":[18],"We":[19,82],"argue":[20],"that,":[21],"while":[22,175],"foundation":[23],"models":[24],"have":[25],"collapsed":[26],"cost":[28,34],"of":[29,35,90],"creating":[30],"these":[31,56],"skills,":[32],"trusting":[36],"them":[37],"has":[38],"not.":[39],"Existing":[40],"skill-evolution":[41],"loops":[42],"refine":[43],"execution":[46],"feedback,":[47],"unit":[48],"tests,":[49],"environment":[50],"reward,":[51],"or":[52,246],"LLM":[53],"self-critique,":[54],"but":[55],"signals":[57],"provide":[58],"only":[59],"trace-level":[60],"evidence:":[61],"they":[62],"show":[63],"that":[64,72,112,130,169,212],"a":[65,85,102,109,127,166],"skill":[66,93,98,142,150,173,235],"worked":[67],"on":[68],"sampled":[69],"executions,":[70],"not":[71],"skill-induced":[73],"plans":[74,146],"satisfy":[75],"temporal":[76,155],"safety":[77],"contracts":[78],"under":[79],"untested":[80],"conditions.":[81],"introduce":[83],"VASO,":[84,96],"framework":[86,211],"for":[87,123,223,232],"verification-guided":[88],"self-evolution":[89],"LLM-generated":[91,221],"contracts.":[94],"In":[95],"each":[97],"is":[99,208],"represented":[100],"as":[101],"semantic":[103],"contract":[104,174],"with":[105,120],"two":[106],"coupled":[107],"interfaces:":[108],"formal":[110,217,227],"interface":[111,129],"aligns":[113],"states,":[115],"observations,":[116],"and":[117,126,153,183,201,219],"control":[118],"commands":[119],"logical":[121],"propositions":[122],"model":[124,136,248],"checking,":[125],"planner-facing":[128],"guides":[131],"executable":[132],"behavior":[133],"generation.":[134],"A":[135],"checker":[137],"first":[138,210],"filters":[139],"logically":[140],"inconsistent":[141],"contracts,":[143,236],"then":[144],"verifies":[145],"induced":[147],"by":[148],"against":[151],"global":[152],"local":[154],"specifications.":[156],"When":[157],"verification":[158,218],"fails,":[159],"VASO":[160,187,207],"translates":[161],"counterexample":[163],"trace":[164],"textual":[167],"gradient":[168],"updates":[170],"reusable":[172,233],"keeping":[176],"foundation-model":[177],"weights":[178],"frozen.":[179],"On":[180],"Clearpath":[181],"Jackal":[182],"PX4":[184],"quadcopter":[185],"tasks,":[186],"reaches":[188],"97.2%":[189],"formal-specification":[190],"compliance":[191],"using":[192],"fewer":[193],"than":[194,238],"100":[195],"optimization":[196,230],"samples,":[197],"outperforming":[198],"execution-feedback,":[199],"prompt-optimization,":[200],"fine-tuning":[202,247],"baselines.":[203],"To":[204],"our":[205],"knowledge,":[206],"closes":[213],"loop":[215],"between":[216],"self-evolving":[220],"AI":[225],"agents:":[226],"counterexamples":[228],"become":[229],"feedback":[231],"rather":[237],"merely":[239],"verifying":[240],"one-off":[241],"plans,":[242],"tuning":[243],"planner":[244],"prompts,":[245],"weights.":[249]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-06T00:00:00"}