{"id":"https://openalex.org/W7155063859","doi":"https://doi.org/10.48550/arxiv.2604.18107","title":"Test-Time Perturbation Learning with Delayed Feedback for Vision-Language-Action Models","display_name":"Test-Time Perturbation Learning with Delayed Feedback for Vision-Language-Action Models","publication_year":2026,"publication_date":"2026-04-20","ids":{"openalex":"https://openalex.org/W7155063859","doi":"https://doi.org/10.48550/arxiv.2604.18107"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.18107","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.18107","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.2604.18107","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114104083","display_name":"Zehua Zang","orcid":"https://orcid.org/0000-0001-8969-3218"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zang, Zehua","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134164791","display_name":"Xi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Xi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134142750","display_name":"Fuchun Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Fuchun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134117868","display_name":"Xiao Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Xiao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134165758","display_name":"Lixiang Lium","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lium, Lixiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134207255","display_name":"Jiahuan Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Jiahuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5134138243","display_name":"Jiangmeng Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Jiangmeng","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":0.8689000010490417,"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":0.8689000010490417,"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/T12026","display_name":"Explainable Artificial Intelligence (XAI)","score":0.021800000220537186,"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":0.01979999989271164,"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/spurious-relationship","display_name":"Spurious relationship","score":0.8271999955177307},{"id":"https://openalex.org/keywords/perturbation","display_name":"Perturbation (astronomy)","score":0.5745000243186951},{"id":"https://openalex.org/keywords/control-theory","display_name":"Control theory (sociology)","score":0.5228999853134155},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.41769999265670776},{"id":"https://openalex.org/keywords/correlation","display_name":"Correlation","score":0.39969998598098755},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.3271999955177307},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.3109000027179718},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.3095000088214874}],"concepts":[{"id":"https://openalex.org/C97256817","wikidata":"https://www.wikidata.org/wiki/Q1462316","display_name":"Spurious relationship","level":2,"score":0.8271999955177307},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6258999705314636},{"id":"https://openalex.org/C177918212","wikidata":"https://www.wikidata.org/wiki/Q803623","display_name":"Perturbation (astronomy)","level":2,"score":0.5745000243186951},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.5228999853134155},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.41769999265670776},{"id":"https://openalex.org/C117220453","wikidata":"https://www.wikidata.org/wiki/Q5172842","display_name":"Correlation","level":2,"score":0.39969998598098755},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3479999899864197},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.3271999955177307},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.3109000027179718},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.3095000088214874},{"id":"https://openalex.org/C51110983","wikidata":"https://www.wikidata.org/wiki/Q16503490","display_name":"Overconfidence effect","level":2,"score":0.3043999969959259},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.2953999936580658},{"id":"https://openalex.org/C2781238097","wikidata":"https://www.wikidata.org/wiki/Q175026","display_name":"Object (grammar)","level":2,"score":0.2939000129699707},{"id":"https://openalex.org/C186886427","wikidata":"https://www.wikidata.org/wiki/Q5441213","display_name":"Feedback loop","level":2,"score":0.289000004529953},{"id":"https://openalex.org/C107464732","wikidata":"https://www.wikidata.org/wiki/Q235781","display_name":"Adaptive control","level":3,"score":0.28439998626708984},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.27630001306533813},{"id":"https://openalex.org/C52970973","wikidata":"https://www.wikidata.org/wiki/Q2497134","display_name":"Adaptive system","level":2,"score":0.2745000123977661},{"id":"https://openalex.org/C2778915421","wikidata":"https://www.wikidata.org/wiki/Q3643177","display_name":"Performance improvement","level":2,"score":0.2662999927997589},{"id":"https://openalex.org/C18020424","wikidata":"https://www.wikidata.org/wiki/Q220858","display_name":"Autopilot","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.2558000087738037},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.2533999979496002},{"id":"https://openalex.org/C25343380","wikidata":"https://www.wikidata.org/wiki/Q277521","display_name":"Relation (database)","level":2,"score":0.2524999976158142},{"id":"https://openalex.org/C81074085","wikidata":"https://www.wikidata.org/wiki/Q366872","display_name":"Motion planning","level":3,"score":0.2508000135421753},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.18107","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.18107","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.2604.18107","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.18107","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":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16","score":0.6508229970932007}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-Language-Action":[0],"models":[1],"(VLAs)":[2],"achieve":[3],"remarkable":[4],"performance":[5,62,89],"in":[6,20,131,150],"sequential":[7],"decision-making":[8,152],"but":[9],"remain":[10],"fragile":[11],"to":[12,27,33,87],"subtle":[13],"environmental":[14],"shifts,":[15],"such":[16],"as":[17],"small":[18],"changes":[19],"object":[21],"pose.":[22],"We":[23,46],"attribute":[24],"this":[25],"brittleness":[26],"trajectory":[28],"overfitting,":[29],"where":[30],"VLAs":[31],"over-attend":[32],"the":[34,65,70],"spurious":[35,71],"correlation":[36,72],"between":[37],"actions":[38],"and":[39,77,90,120,137],"entities,":[40],"then":[41],"reproduce":[42],"memorized":[43],"action":[44,78,105],"patterns.":[45],"propose":[47],"Perturbation":[48],"learning":[49],"with":[50,139],"Delayed":[51],"Feedback":[52],"(PDF),":[53],"a":[54,98,143],"verifier-free":[55],"test-time":[56,140,148],"adaptation":[57,149],"framework":[58],"that":[59,102],"improves":[60],"decision":[61],"without":[63],"fine-tuning":[64],"base":[66],"model.":[67],"PDF":[68,96,130],"mitigates":[69],"through":[73],"uncertainty-based":[74],"data":[75],"augmentation":[76,85],"voting,":[79],"while":[80],"an":[81],"adaptive":[82],"scheduler":[83],"allocates":[84],"budgets":[86],"balance":[88],"efficiency.":[91],"To":[92],"further":[93],"improve":[94],"stability,":[95],"learns":[97],"lightweight":[99],"perturbation":[100],"module":[101],"retrospectively":[103],"adjusts":[104],"logits":[106],"guided":[107],"by":[108],"delayed":[109],"feedback,":[110],"correcting":[111],"overconfidence":[112],"issue.":[113],"Experiments":[114],"on":[115],"LIBERO":[116],"(+7.4\\%":[117],"success":[118,133],"rate)":[119],"Atari":[121],"(+10.3":[122],"human":[123],"normalized":[124],"score)":[125],"demonstrate":[126],"consistent":[127],"gains":[128],"of":[129],"task":[132],"over":[134],"vanilla":[135],"VLA":[136,138],"adaptation,":[141],"establishing":[142],"practical":[144],"path":[145],"toward":[146],"reliable":[147],"multimodal":[151],"agents.":[153],"The":[154],"code":[155],"is":[156],"available":[157],"at":[158],"\\href{https://github.com/zhoujiahuan1991/CVPR2026-PDF}{https://github.com/zhoujiahuan1991/CVPR2026-PDF}.":[159]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-04-22T00:00:00"}