{"id":"https://openalex.org/W7128494407","doi":"https://doi.org/10.48550/arxiv.2602.07845","title":"Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoning","display_name":"Recurrent-Depth VLA: Implicit Test-Time Compute Scaling of Vision-Language-Action Models via Latent Iterative Reasoning","publication_year":2026,"publication_date":"2026-02-08","ids":{"openalex":"https://openalex.org/W7128494407","doi":"https://doi.org/10.48550/arxiv.2602.07845"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.07845","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5075380440","display_name":"Yalcin Tur","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Tur, Yalcin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125594629","display_name":"Jalal Naghiyev","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Naghiyev, Jalal","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5085879597","display_name":"Haoquan Fang","orcid":"https://orcid.org/0009-0007-9759-3867"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fang, Haoquan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125489104","display_name":"Wei-Chuan Tsai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tsai, Wei-Chuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027184297","display_name":"Jiafei Duan","orcid":"https://orcid.org/0000-0003-4551-2424"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Duan, Jiafei","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5031489518","display_name":"Dieter Fox","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fox, Dieter","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5125485704","display_name":"Ranjay Krishna","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Krishna, Ranjay","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5075380440"],"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":0.957099974155426,"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":0.957099974155426,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.006599999964237213,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.00419999985024333,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6700999736785889},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5777999758720398},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5598999857902527},{"id":"https://openalex.org/keywords/path","display_name":"Path (computing)","score":0.4887000024318695},{"id":"https://openalex.org/keywords/latent-variable","display_name":"Latent variable","score":0.486299991607666},{"id":"https://openalex.org/keywords/constant","display_name":"Constant (computer programming)","score":0.4814999997615814},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.4577000141143799},{"id":"https://openalex.org/keywords/simple","display_name":"Simple (philosophy)","score":0.44909998774528503}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.734499990940094},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6700999736785889},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5777999758720398},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5598999857902527},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.4887000024318695},{"id":"https://openalex.org/C51167844","wikidata":"https://www.wikidata.org/wiki/Q4422623","display_name":"Latent variable","level":2,"score":0.486299991607666},{"id":"https://openalex.org/C2777027219","wikidata":"https://www.wikidata.org/wiki/Q1284190","display_name":"Constant (computer programming)","level":2,"score":0.4814999997615814},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.4577000141143799},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.4544999897480011},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.453000009059906},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.44909998774528503},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.4293000102043152},{"id":"https://openalex.org/C182365436","wikidata":"https://www.wikidata.org/wiki/Q50701","display_name":"Variable (mathematics)","level":2,"score":0.40049999952316284},{"id":"https://openalex.org/C155032097","wikidata":"https://www.wikidata.org/wiki/Q798503","display_name":"Backpropagation","level":3,"score":0.37549999356269836},{"id":"https://openalex.org/C65965080","wikidata":"https://www.wikidata.org/wiki/Q1806885","display_name":"Latent variable model","level":3,"score":0.37220001220703125},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.37220001220703125},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3400000035762787},{"id":"https://openalex.org/C2780069185","wikidata":"https://www.wikidata.org/wiki/Q7977945","display_name":"Equivalence (formal languages)","level":2,"score":0.3208000063896179},{"id":"https://openalex.org/C159694833","wikidata":"https://www.wikidata.org/wiki/Q2321565","display_name":"Iterative method","level":2,"score":0.2930000126361847},{"id":"https://openalex.org/C2779982483","wikidata":"https://www.wikidata.org/wiki/Q6094420","display_name":"Iterative refinement","level":2,"score":0.2833000123500824},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.27880001068115234},{"id":"https://openalex.org/C5465570","wikidata":"https://www.wikidata.org/wiki/Q5326898","display_name":"Early stopping","level":3,"score":0.2784999907016754},{"id":"https://openalex.org/C148220186","wikidata":"https://www.wikidata.org/wiki/Q7111912","display_name":"Outcome (game theory)","level":2,"score":0.2711000144481659},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.26440000534057617},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.2524000108242035}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.07845","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.07845","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.07845","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":"article"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2602.07845","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"publisher-specific-oa","license_id":"https://openalex.org/licenses/publisher-specific-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"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":{"Current":[0],"Vision-Language-Action":[1],"(VLA)":[2],"models":[3],"rely":[4],"on":[5,15,105,109],"fixed":[6],"computational":[7,49],"depth,":[8],"expending":[9],"the":[10,90],"same":[11],"amount":[12],"of":[13],"compute":[14,98,148],"simple":[16],"adjustments":[17],"and":[18,33,162],"complex":[19],"multi-step":[20],"manipulation.":[21],"While":[22],"Chain-of-Thought":[23],"(CoT)":[24],"prompting":[25],"enables":[26],"variable":[27],"computation,":[28],"it":[29],"scales":[30],"memory":[31,75,160],"linearly":[32],"is":[34,79,117],"ill-suited":[35],"for":[36],"continuous":[37],"action":[38,65],"spaces.":[39],"We":[40],"introduce":[41],"Recurrent-Depth":[42],"VLA":[43,171],"(RD-VLA),":[44],"an":[45,100],"architecture":[46],"that":[47,67,114,120],"achieves":[48],"adaptivity":[50],"via":[51],"latent":[52,106,155],"iterative":[53],"refinement":[54,91],"rather":[55],"than":[56],"explicit":[57],"token":[58],"generation.":[59],"RD-VLA":[60,95,141],"employs":[61],"a":[62,73,143],"recurrent,":[63],"weight-tied":[64],"head":[66],"supports":[68],"arbitrary":[69],"inference":[70,128,166],"depth":[71,116],"with":[72,126,133,154],"constant":[74,159],"footprint.":[76],"The":[77],"model":[78],"trained":[80],"using":[81,99],"truncated":[82],"backpropagation":[83],"through":[84],"time":[85],"(TBPTT)":[86],"to":[87,146,157,164],"efficiently":[88],"supervise":[89],"process.":[92],"At":[93],"inference,":[94],"dynamically":[96],"allocates":[97],"adaptive":[101],"stopping":[102],"criterion":[103],"based":[104],"convergence.":[107],"Experiments":[108],"challenging":[110],"manipulation":[111],"tasks":[112,119,138],"show":[113],"recurrent":[115],"critical:":[118],"fail":[121],"entirely":[122],"(0":[123],"percent":[124,131],"success)":[125],"single-iteration":[127],"exceed":[129],"90":[130],"success":[132],"four":[134],"iterations,":[135],"while":[136],"simpler":[137],"saturate":[139],"rapidly.":[140],"provides":[142],"scalable":[144],"path":[145],"test-time":[147],"in":[149],"robotics,":[150],"replacing":[151],"token-based":[152],"reasoning":[153,156],"achieve":[158],"usage":[161],"up":[163],"80x":[165],"speedup":[167],"over":[168],"prior":[169],"reasoning-based":[170],"models.":[172],"Project":[173],"page:":[174],"https://rd-vla.github.io/":[175]},"counts_by_year":[],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2026-02-11T00:00:00"}