{"id":"https://openalex.org/W7164407716","doi":"https://doi.org/10.48550/arxiv.2606.12402","title":"DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?","display_name":"DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?","publication_year":2026,"publication_date":"2026-06-10","ids":{"openalex":"https://openalex.org/W7164407716","doi":"https://doi.org/10.48550/arxiv.2606.12402"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.12402","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.12402","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.2606.12402","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114988566","display_name":"Jadelynn Dao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Dao, Jadelynn","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138424330","display_name":"Milan Ganai","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ganai, Milan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138448817","display_name":"Yasmina Abukhadra","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Abukhadra, Yasmina","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5004772595","display_name":"Ajay Sridhar","orcid":"https://orcid.org/0009-0008-2268-6097"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sridhar, Ajay","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016289526","display_name":"Mozhgan Nasr Azadani","orcid":"https://orcid.org/0000-0002-2872-1438"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Azadani, Mozhgan Nasr","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138404517","display_name":"Katie Luo","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Luo, Katie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138412736","display_name":"Clark Barrett","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Barrett, Clark","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138412213","display_name":"Jiajun Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Jiajun","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138471461","display_name":"Chelsea Finn","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Finn, Chelsea","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5138464635","display_name":"Marco Pavone","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pavone, Marco","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":0.38280001282691956,"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":0.38280001282691956,"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":0.13860000669956207,"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/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.09449999779462814,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.6632999777793884},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6353999972343445},{"id":"https://openalex.org/keywords/embodied-cognition","display_name":"Embodied cognition","score":0.6326000094413757},{"id":"https://openalex.org/keywords/fraction","display_name":"Fraction (chemistry)","score":0.44200000166893005},{"id":"https://openalex.org/keywords/limiting","display_name":"Limiting","score":0.38510000705718994},{"id":"https://openalex.org/keywords/router","display_name":"Router","score":0.36970001459121704},{"id":"https://openalex.org/keywords/component","display_name":"Component (thermodynamics)","score":0.3684999942779541},{"id":"https://openalex.org/keywords/pareto-principle","display_name":"Pareto principle","score":0.36570000648498535}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6766999959945679},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.6632999777793884},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6353999972343445},{"id":"https://openalex.org/C100609095","wikidata":"https://www.wikidata.org/wiki/Q1335050","display_name":"Embodied cognition","level":2,"score":0.6326000094413757},{"id":"https://openalex.org/C149629883","wikidata":"https://www.wikidata.org/wiki/Q660926","display_name":"Fraction (chemistry)","level":2,"score":0.44200000166893005},{"id":"https://openalex.org/C188198153","wikidata":"https://www.wikidata.org/wiki/Q1613840","display_name":"Limiting","level":2,"score":0.38510000705718994},{"id":"https://openalex.org/C2775896111","wikidata":"https://www.wikidata.org/wiki/Q642560","display_name":"Router","level":2,"score":0.36970001459121704},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.3684999942779541},{"id":"https://openalex.org/C137635306","wikidata":"https://www.wikidata.org/wiki/Q182667","display_name":"Pareto principle","level":2,"score":0.36570000648498535},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.32829999923706055},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.3093000054359436},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.30230000615119934},{"id":"https://openalex.org/C189645446","wikidata":"https://www.wikidata.org/wiki/Q350865","display_name":"Mirroring","level":2,"score":0.2953000068664551},{"id":"https://openalex.org/C2778571376","wikidata":"https://www.wikidata.org/wiki/Q1355821","display_name":"Frontier","level":2,"score":0.2928999960422516},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.29010000824928284},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.28940001130104065},{"id":"https://openalex.org/C51485801","wikidata":"https://www.wikidata.org/wiki/Q16966861","display_name":"Efficient frontier","level":3,"score":0.28769999742507935},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.2759000062942505},{"id":"https://openalex.org/C14036430","wikidata":"https://www.wikidata.org/wiki/Q3736076","display_name":"Function (biology)","level":2,"score":0.2653999924659729},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.26460000872612},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.2614000141620636},{"id":"https://openalex.org/C28719098","wikidata":"https://www.wikidata.org/wiki/Q44946","display_name":"Point (geometry)","level":2,"score":0.25189998745918274}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.12402","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.12402","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.2606.12402","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.12402","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":{"Vision-Language":[0],"Models":[1],"(VLMs)":[2],"are":[3],"increasingly":[4],"deployed":[5],"as":[6],"high-level":[7],"planners":[8],"for":[9],"embodied":[10,46,186],"agents,":[11],"with":[12],"an":[13],"emerging":[14],"strategy":[15],"of":[16,194],"scaling":[17,100,175],"test-time":[18,60,118,176],"compute":[19,61,85,119,177],"to":[20,58,64,68,83,164],"improve":[21],"capability.":[22],"However,":[23],"we":[24],"observe":[25],"that":[26,53,78,117,173,181],"doing":[27],"so":[28],"increases":[29],"latency,":[30],"token":[31],"usage,":[32],"and":[33,56,107,114,148,180],"FLOPs":[34],"while":[35],"yielding":[36],"uneven,":[37],"often":[38],"diminishing":[39],"gains":[40],"in":[41,141,188],"downstream":[42],"success,":[43],"limiting":[44],"where":[45,57,151],"agents":[47],"can":[48,183,199],"be":[49,200],"deployed.":[50],"We":[51,72,132],"argue":[52],"choosing":[54],"when":[55],"spend":[59],"is":[62,120,178],"central":[63],"bringing":[65],"frontier":[66,92],"performance":[67],"the":[69,89,195],"real":[70],"world.":[71],"introduce":[73],"DIRECT,":[74],"a":[75,122,137,142,157,192],"routing":[76],"framework":[77],"uses":[79],"multimodal":[80],"scene":[81],"context":[82],"allocate":[84],"per":[86],"prompt,":[87],"improving":[88],"success--cost":[90],"Pareto":[91],"over":[93],"fixed":[94],"model":[95,105],"selection.":[96],"Across":[97],"three":[98],"dominant":[99],"axes,":[101],"namely":[102],"chain-of-thought":[103],"depth,":[104],"size,":[106],"memory":[108],"history,":[109],"our":[110,152,170],"experiments":[111],"on":[112,136],"VLABench":[113],"RoboMME":[115],"show":[116,172],"not":[121],"uniform":[123],"lever:":[124],"different":[125],"axes":[126],"yield":[127],"qualitatively":[128],"distinct":[129],"capability":[130],"gains.":[131],"validate":[133],"these":[134],"insights":[135],"physical":[138],"Franka":[139],"arm":[140],"DROID":[143],"setup":[144],"spanning":[145],"zero-shot":[146],"manipulation":[147],"long-horizon":[149],"chaining,":[150],"router":[153],"matches":[154],"or":[155],"exceeds":[156],"stronger":[158],"model's":[159],"success":[160],"rate":[161],"at":[162,191,202],"up":[163],"65%":[165],"lower":[166],"average":[167],"latency.":[168],"Ultimately,":[169],"results":[171],"naively":[174],"wasteful,":[179],"DIRECT":[182],"provide":[184],"frontier-level":[185],"planning":[187],"robotic":[189],"systems":[190],"fraction":[193],"cost.":[196],"Project":[197],"page":[198],"found":[201],"jadee-dao.github.io/direct/.":[203]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-12T00:00:00"}