{"id":"https://openalex.org/W7214700375","doi":"https://doi.org/10.48550/arxiv.2609.33256","title":"ActionGround: Training-Free Runtime Refinement of Frozen VLA Policies","display_name":"ActionGround: Training-Free Runtime Refinement of Frozen VLA Policies","publication_year":2026,"publication_date":"2026-09-27","ids":{"openalex":"https://openalex.org/W7214700375","doi":"https://doi.org/10.48550/arxiv.2609.33256"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2609.33256","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.33256","pdf_url":"https://arxiv.org/pdf/2609.33256","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","is_traditional":true,"indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2609.33256","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129692628","display_name":"Namai Chandra","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I24676775","display_name":"Indian Institute of Technology Madras","ror":"https://ror.org/03v0r5n49","country_code":"IN","type":"education","lineage":["https://openalex.org/I24676775"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Chandra, Namai","raw_affiliation_strings":["Indian Institute of Technology Madras"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Indian Institute of Technology Madras","institution_ids":["https://openalex.org/I24676775"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5140889595","display_name":"Madhur Thareja","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I24676775","display_name":"Indian Institute of Technology Madras","ror":"https://ror.org/03v0r5n49","country_code":"IN","type":"education","lineage":["https://openalex.org/I24676775"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Thareja, Madhur","raw_affiliation_strings":["Indian AI Research Organisation","Indian Institute of Technology Madras"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Indian AI Research Organisation","institution_ids":[]},{"raw_affiliation_string":"Indian Institute of Technology Madras","institution_ids":["https://openalex.org/I24676775"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5119525110","display_name":"Shriram Damodaran","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Damodaran, Shriram","raw_affiliation_strings":["Nanyang Technological University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University","institution_ids":["https://openalex.org/I172675005"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5148936431","display_name":"Addison Lin Wang","orcid":null,"observed_orcids":[]},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Wang, Addison Lin","raw_affiliation_strings":["Nanyang Technological University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanyang Technological University","institution_ids":["https://openalex.org/I172675005"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":null,"has_fulltext":true,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.9940999746322632,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T10653","display_name":"Robot Manipulation and Learning","score":0.9940999746322632,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems 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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.005200000014156103,"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":9.999999747378752e-05,"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/vision-language-action-models","display_name":"vision-language-action models","score":0.906000018119812},{"id":"https://openalex.org/keywords/euler-lagrange-dynamics","display_name":"Euler-Lagrange dynamics","score":0.8450000286102295},{"id":"https://openalex.org/keywords/robotic-manipulation","display_name":"robotic manipulation","score":0.5559999942779541},{"id":"https://openalex.org/keywords/pick-and-place","display_name":"pick-and-place","score":0.9200000166893005},{"id":"https://openalex.org/keywords/franka-emika-panda","display_name":"Franka Emika Panda","score":0.6069999933242798},{"id":"https://openalex.org/keywords/trajectory-stability","display_name":"trajectory stability","score":0.5210000276565552}],"concepts":[{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.6829000115394592},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6564000248908997},{"id":"https://openalex.org/C175291020","wikidata":"https://www.wikidata.org/wiki/Q1156822","display_name":"Offset (computer science)","level":2,"score":0.59170001745224},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.5605999827384949},{"id":"https://openalex.org/C61445026","wikidata":"https://www.wikidata.org/wiki/Q217608","display_name":"Fixed point","level":2,"score":0.5101000070571899},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.46619999408721924},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4593999981880188},{"id":"https://openalex.org/C2776436953","wikidata":"https://www.wikidata.org/wiki/Q5163215","display_name":"Consistency (knowledge bases)","level":2,"score":0.42149999737739563},{"id":"https://openalex.org/C2780154230","wikidata":"https://www.wikidata.org/wiki/Q513420","display_name":"Undo","level":2,"score":0.40119999647140503},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3984000086784363},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.39430001378059387},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.38679999113082886},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.38499999046325684},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.38370001316070557},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.3375999927520752},{"id":"https://openalex.org/C2779960059","wikidata":"https://www.wikidata.org/wiki/Q7113681","display_name":"Overhead (engineering)","level":2,"score":0.33660000562667847},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.32850000262260437},{"id":"https://openalex.org/C77405623","wikidata":"https://www.wikidata.org/wiki/Q598451","display_name":"System dynamics","level":2,"score":0.31450000405311584},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.30820000171661377},{"id":"https://openalex.org/C165838908","wikidata":"https://www.wikidata.org/wiki/Q736777","display_name":"Calibration","level":2,"score":0.3075999915599823},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.30379998683929443},{"id":"https://openalex.org/C61797465","wikidata":"https://www.wikidata.org/wiki/Q1188986","display_name":"Term (time)","level":2,"score":0.2870999872684479},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.28690001368522644},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.28439998626708984},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.27720001339912415},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.25949999690055847},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.2565999925136566},{"id":"https://openalex.org/C17500928","wikidata":"https://www.wikidata.org/wiki/Q959968","display_name":"Control system","level":2,"score":0.2549999952316284},{"id":"https://openalex.org/C52723943","wikidata":"https://www.wikidata.org/wiki/Q1127410","display_name":"Serialization","level":2,"score":0.2531000077724457}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2609.33256","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.33256","pdf_url":"https://arxiv.org/pdf/2609.33256","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2609.33256","doi":"https://doi.org/10.48550/arxiv.2609.33256","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2609.33256","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,"listed_in":[],"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"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2609.33256","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2609.33256","pdf_url":"https://arxiv.org/pdf/2609.33256","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7214700375.pdf","grobid_xml":"https://content.openalex.org/works/W7214700375.grobid-xml"},"referenced_works_count":14,"referenced_works":["https://openalex.org/W2783192199","https://openalex.org/W3036143115","https://openalex.org/W3088304681","https://openalex.org/W3113149630","https://openalex.org/W4385430679","https://openalex.org/W4399695759","https://openalex.org/W4405031329","https://openalex.org/W4405090228","https://openalex.org/W4409147637","https://openalex.org/W4414079054","https://openalex.org/W4414898213","https://openalex.org/W4416095924","https://openalex.org/W6948104436","https://openalex.org/W7138212329"],"related_works":["https://openalex.org/W7164812032","https://openalex.org/W7134965772","https://openalex.org/W7166541883","https://openalex.org/W7220671413","https://openalex.org/W7214048529","https://openalex.org/W7081569152","https://openalex.org/W7167900256","https://openalex.org/W7155090762","https://openalex.org/W7125931475","https://openalex.org/W7083697555"],"abstract_inverted_index":{"Vision-Language-Action":[0],"(VLA)":[1],"models":[2],"map":[3],"visual":[4],"observations":[5],"and":[6,75,121,138,150],"language":[7],"instructions":[8],"directly":[9],"to":[10,146,154,164,191,203,231],"robot":[11],"actions,":[12],"but":[13],"they":[14],"do":[15],"not":[16],"explicitly":[17],"represent":[18],"the":[19,26,67,89,219],"phase":[20,69],"structure":[21],"of":[22,56,92],"manipulation":[23,68],"tasks":[24,127,137],"or":[25,48,73],"rigid-body":[27],"dynamics":[28,100],"governing":[29],"execution.":[30],"We":[31,114],"present":[32],"ActionGround,":[33],"a":[34,41,77,105,112,129,167,175,186,192,211],"neuro-symbolic,":[35],"training-free":[36],"runtime":[37],"layer":[38],"that":[39],"wraps":[40],"frozen":[42],"VLA":[43],"policy":[44],"without":[45],"retraining,":[46],"fine-tuning,":[47],"weight":[49],"access,":[50],"adding":[51],"less":[52],"than":[53,109],"1":[54],"ms":[55],"overhead":[57],"per":[58],"control":[59,96],"step.":[60],"A":[61],"symbolic":[62],"phase-aware":[63],"finite-state":[64],"machine":[65],"identifies":[66],"(approach,":[70],"grasp,":[71],"transport,":[72],"place)":[74],"applies":[76],"phase-specific":[78],"rule-based":[79],"correction.":[80],"In":[81,166,185],"parallel,":[82],"an":[83],"always-on,":[84],"inertia-weighted":[85],"Euler-Lagrange":[86],"term":[87],"incorporates":[88],"robot's":[90],"equations":[91],"motion":[93],"into":[94],"each":[95],"step,":[97],"while":[98,158],"its":[99],"residual":[101],"is":[102,208,222,229],"logged":[103],"as":[104,111,210],"consistency":[106],"diagnostic":[107],"rather":[108],"used":[110],"gate.":[113],"evaluate":[115],"ActionGround":[116,140,172],"across":[117,136],"OpenVLA,":[118],"OpenVLA-OFT,":[119],"Force-VLA,":[120],"Generalist-VLA":[122],"on":[123,218],"ten":[124],"LIBERO-Spatial":[125],"pick-and-place":[126,233],"using":[128],"7-DoF":[130],"Franka":[131],"Panda.":[132],"With":[133],"fixed":[134],"parameters":[135],"backbones,":[139],"improves":[141],"success":[142,199,217],"rate":[143],"by":[144,152,162],"up":[145,153,163],"6":[147],"percentage":[148,156],"points":[149],"stability":[151],"19.3":[155],"points,":[157],"improving":[159],"trajectory":[160],"efficiency":[161],"15%.":[165],"separate":[168],"Robosuite":[169,188],"noise":[170],"sweep,":[171],"provides":[173],"approximately":[174],"10x":[176],"improvement":[177],"in":[178],"trajectory-jerk":[179],"robustness":[180],"under":[181],"injected":[182],"action":[183],"noise.":[184],"matched-seed":[187],"simulation":[189],"companion":[190],"real":[193,220],"Agilex":[194],"Piper":[195],"trial,":[196],"simulated":[197],"baseline":[198],"increases":[200],"from":[201],"35%":[202],"95%.":[204],"The":[205],"physical-hardware":[206],"experiment":[207],"presented":[209],"qualitative":[212],"deployment":[213],"demonstration;":[214],"quantitative":[215],"per-trial":[216],"arm":[221],"left":[223],"for":[224],"future":[225],"work.":[226],"Our":[227],"evaluation":[228],"limited":[230],"rigid-object":[232],"manipulation.":[234]},"counts_by_year":[],"updated_date":"2026-11-10T11:39:43.021295","created_date":"2026-09-30T00:00:00"}