{"id":"https://openalex.org/W7166674768","doi":"https://doi.org/10.48550/arxiv.2606.29892","title":"Trust Your Instincts: Confidence-Driven Test-Time RL for Vision-Language-Action Models","display_name":"Trust Your Instincts: Confidence-Driven Test-Time RL for Vision-Language-Action Models","publication_year":2026,"publication_date":"2026-06-29","ids":{"openalex":"https://openalex.org/W7166674768","doi":"https://doi.org/10.48550/arxiv.2606.29892"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.29892","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.29892","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.2606.29892","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139690219","display_name":"Siyao Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Siyao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139699846","display_name":"Jiakang Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yuan, Jiakang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139649893","display_name":"Jiaxin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jiaxin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5050105433","display_name":"T X Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Tao","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":0.6953999996185303,"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":0.6953999996185303,"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":0.15479999780654907,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.03720000013709068,"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/reinforcement-learning","display_name":"Reinforcement learning","score":0.6933000087738037},{"id":"https://openalex.org/keywords/bootstrapping","display_name":"Bootstrapping (finance)","score":0.5630000233650208},{"id":"https://openalex.org/keywords/oracle","display_name":"Oracle","score":0.5611000061035156},{"id":"https://openalex.org/keywords/similarity","display_name":"Similarity (geometry)","score":0.3547999858856201},{"id":"https://openalex.org/keywords/dual","display_name":"Dual (grammatical number)","score":0.3499999940395355},{"id":"https://openalex.org/keywords/supervised-learning","display_name":"Supervised learning","score":0.32010000944137573}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7810999751091003},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.6933000087738037},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.6035000085830688},{"id":"https://openalex.org/C207609745","wikidata":"https://www.wikidata.org/wiki/Q4944086","display_name":"Bootstrapping (finance)","level":2,"score":0.5630000233650208},{"id":"https://openalex.org/C55166926","wikidata":"https://www.wikidata.org/wiki/Q2892946","display_name":"Oracle","level":2,"score":0.5611000061035156},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.517799973487854},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.3547999858856201},{"id":"https://openalex.org/C2780980858","wikidata":"https://www.wikidata.org/wiki/Q110022","display_name":"Dual (grammatical number)","level":2,"score":0.3499999940395355},{"id":"https://openalex.org/C136389625","wikidata":"https://www.wikidata.org/wiki/Q334384","display_name":"Supervised learning","level":3,"score":0.32010000944137573},{"id":"https://openalex.org/C126388530","wikidata":"https://www.wikidata.org/wiki/Q1131737","display_name":"Imitation","level":2,"score":0.3012999892234802},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.2632000148296356},{"id":"https://openalex.org/C4679612","wikidata":"https://www.wikidata.org/wiki/Q866298","display_name":"Aggregate (composite)","level":2,"score":0.2615000009536743},{"id":"https://openalex.org/C2779436431","wikidata":"https://www.wikidata.org/wiki/Q30672407","display_name":"Policy learning","level":2,"score":0.25920000672340393},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.2524000108242035}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.29892","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.29892","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.2606.29892","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.29892","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":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Reinforcement":[0],"learning":[1],"(RL)":[2],"has":[3],"become":[4],"indispensable":[5],"for":[6,118,125],"pushing":[7],"Vision-Language-Action":[8],"Models":[9],"(VLAs)":[10],"beyond":[11],"static":[12],"imitation":[13],"learning.":[14],"However,":[15],"existing":[16],"RL":[17,72,146],"methods":[18],"typically":[19],"require":[20],"external":[21,87,155],"environmental":[22],"feedback,":[23],"relying":[24,85],"on":[25,61,86,130],"predefined":[26],"success":[27],"signals":[28],"to":[29,58,78,93,161],"guide":[30],"policy":[31,81],"updates.":[32],"In":[33,102],"this":[34,62],"work,":[35],"we":[36,64,104],"show":[37,136],"that":[38,74,137],"VLA":[39,76,163],"models":[40,77],"possess":[41],"useful":[42],"internal":[43],"evaluative":[44],"capabilities:":[45],"in":[46],"discrete-action":[47],"VLAs,":[48],"trajectories":[49],"with":[50,148],"higher":[51],"generation":[52],"confidence":[53],"are":[54],"significantly":[55],"more":[56],"likely":[57],"succeed.":[59],"Based":[60],"observation,":[63],"introduce":[65],"T^2VLA":[66,89,138,159],"(Test-time":[67],"VLA),":[68],"an":[69,98],"architecture-agnostic":[70],"test-time":[71],"framework":[73],"enables":[75],"achieve":[79],"self-bootstrapping":[80],"improvement.":[82],"Instead":[83],"of":[84],"rewards,":[88,150],"leverages":[90],"trajectory-level":[91],"similarity":[92],"high-confidence":[94],"expert":[95],"demonstrations":[96],"as":[97],"intrinsic":[99],"reward":[100,156],"signal.":[101],"addition,":[103],"propose":[105],"a":[106,115,121],"Confidence-Driven":[107],"Dual":[108],"Expert":[109,123],"Bootstrapping":[110],"mechanism,":[111],"which":[112],"dynamically":[113],"balances":[114],"Local":[116],"Pseudo-Expert":[117],"exploration":[119],"and":[120,133,143,168],"Global":[122],"Pool":[124],"training":[126],"stability.":[127],"Extensive":[128],"experiments":[129],"the":[131,169],"LIBERO":[132],"RoboTwin":[134],"benchmarks":[135],"consistently":[139],"outperforms":[140],"supervised":[141],"baselines":[142],"approaches":[144],"oracle":[145],"performance":[147],"ground-truth":[149],"achieving":[151],"effective":[152],"improvement":[153],"without":[154],"feedback.":[157],"Furthermore,":[158],"adapts":[160],"distinct":[162],"paradigms,":[164],"including":[165],"both":[166],"OpenVLA-OFT":[167],"pi":[170],"series.":[171]},"counts_by_year":[],"updated_date":"2026-07-01T06:29:00.853634","created_date":"2026-07-01T00:00:00"}