{"id":"https://openalex.org/W7131882670","doi":"https://doi.org/10.48550/arxiv.2602.22474","title":"When to Act, Ask, or Learn: Uncertainty-Aware Policy Steering","display_name":"When to Act, Ask, or Learn: Uncertainty-Aware Policy Steering","publication_year":2026,"publication_date":"2026-02-25","ids":{"openalex":"https://openalex.org/W7131882670","doi":"https://doi.org/10.48550/arxiv.2602.22474"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.22474","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/A5127078209","display_name":"Jessie Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Yuan, Jessie","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5102372129","display_name":"Yilin Wu","orcid":"https://orcid.org/0009-0007-0296-550X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Yilin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5050279893","display_name":"Andrea Bajcsy","orcid":"https://orcid.org/0000-0001-7969-9376"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bajcsy, Andrea","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5127078209"],"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.26840001344680786,"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"}},"topics":[{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.26840001344680786,"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":0.22750000655651093,"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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.2013999968767166,"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/ambiguity","display_name":"Ambiguity","score":0.7519000172615051},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.6240000128746033},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.6195999979972839},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6086999773979187},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.4156000018119812},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.41269999742507935},{"id":"https://openalex.org/keywords/natural-language","display_name":"Natural language","score":0.3580999970436096},{"id":"https://openalex.org/keywords/knowledge-base","display_name":"Knowledge base","score":0.35350000858306885}],"concepts":[{"id":"https://openalex.org/C2780522230","wikidata":"https://www.wikidata.org/wiki/Q1140419","display_name":"Ambiguity","level":2,"score":0.7519000172615051},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.744700014591217},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.6240000128746033},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.6195999979972839},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6086999773979187},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4986000061035156},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4408000111579895},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.4156000018119812},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.41269999742507935},{"id":"https://openalex.org/C195324797","wikidata":"https://www.wikidata.org/wiki/Q33742","display_name":"Natural language","level":2,"score":0.3580999970436096},{"id":"https://openalex.org/C4554734","wikidata":"https://www.wikidata.org/wiki/Q593744","display_name":"Knowledge base","level":2,"score":0.35350000858306885},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.31279999017715454},{"id":"https://openalex.org/C155512373","wikidata":"https://www.wikidata.org/wiki/Q287450","display_name":"Residual","level":2,"score":0.3084000051021576},{"id":"https://openalex.org/C2779439875","wikidata":"https://www.wikidata.org/wiki/Q1078276","display_name":"Natural language understanding","level":3,"score":0.303600013256073},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3010999858379364},{"id":"https://openalex.org/C2779436431","wikidata":"https://www.wikidata.org/wiki/Q30672407","display_name":"Policy learning","level":2,"score":0.2797999978065491},{"id":"https://openalex.org/C114289077","wikidata":"https://www.wikidata.org/wiki/Q3284399","display_name":"Statistical model","level":2,"score":0.2653000056743622},{"id":"https://openalex.org/C112930515","wikidata":"https://www.wikidata.org/wiki/Q4389547","display_name":"Risk analysis (engineering)","level":1,"score":0.2644999921321869},{"id":"https://openalex.org/C2776608160","wikidata":"https://www.wikidata.org/wiki/Q4785462","display_name":"Natural (archaeology)","level":2,"score":0.2597000002861023},{"id":"https://openalex.org/C90329073","wikidata":"https://www.wikidata.org/wiki/Q914232","display_name":"Ask price","level":2,"score":0.2531999945640564},{"id":"https://openalex.org/C165696696","wikidata":"https://www.wikidata.org/wiki/Q11287","display_name":"Exploit","level":2,"score":0.25270000100135803}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.22474","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.2602.22474","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.22474","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.2602.22474","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":[{"id":"https://metadata.un.org/sdg/16","score":0.44900256395339966,"display_name":"Peace, Justice and strong institutions"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Policy":[0],"steering":[1,66,90],"is":[2,134],"an":[3,107],"emerging":[4],"way":[5],"to":[6,44,127,144,175,186,222],"adapt":[7],"robot":[8],"behaviors":[9],"at":[10,137,237],"deployment-time:":[11],"a":[12,21,92,112],"learned":[13],"verifier":[14,161],"analyzes":[15],"low-level":[16,77,102,130],"action":[17,78,103,125],"samples":[18],"proposed":[19],"by":[20],"pre-trained":[22,84,153,181],"policy":[23,89,131],"(e.g.,":[24],"diffusion":[25],"policy)":[26],"and":[27,76,101,105,151,203,213,216,225],"selects":[28,106,162],"only":[29],"those":[30],"aligned":[31],"with":[32,190],"the":[33,58,65,83,129,138,146,149,152,160,163,177,180,184],"task.":[34,139],"While":[35],"Vision-Language":[36],"Models":[37],"(VLMs)":[38],"are":[39,54],"promising":[40],"general-purpose":[41],"verifiers":[42],"due":[43],"their":[45],"reasoning":[46],"capabilities,":[47],"existing":[48],"frameworks":[49],"often":[50],"assume":[51],"these":[52],"models":[53],"well-calibrated.":[55],"In":[56],"practice,":[57],"overconfident":[59],"judgment":[60],"from":[61],"VLM":[62,150],"can":[63,209,234],"degrade":[64],"performance":[67],"under":[68],"both":[69],"high-level":[70],"semantic":[71,98],"uncertainty":[72,79,100,108],"in":[73,201],"task":[74,99,116],"specifications":[75],"or":[80,122,228],"incapability":[81],"of":[82,148,179],"policy.":[85],"We":[86,140,195],"propose":[87],"uncertainty-aware":[88],"(UPS),":[91],"framework":[93,198],"that":[94,159,207],"jointly":[95],"reasons":[96],"about":[97],"feasibility,":[104],"resolution":[109],"strategy:":[110],"execute":[111],"high-confidence":[113],"action,":[114],"clarify":[115],"ambiguity":[117],"via":[118],"natural":[119],"language":[120],"queries,":[121],"ask":[123],"for":[124],"interventions":[126,168,220],"correct":[128,164],"when":[132],"it":[133],"deemed":[135],"incapable":[136,214],"leverage":[141],"conformal":[142],"prediction":[143],"calibrate":[145],"composition":[147],"base":[154],"policy,":[155,182],"providing":[156],"statistical":[157],"assurances":[158],"strategy.":[165],"After":[166],"collecting":[167],"during":[169],"deployment,":[170],"we":[171],"employ":[172],"residual":[173],"learning":[174,231],"improve":[176],"capability":[178],"enabling":[183],"system":[185],"learn":[187],"continually":[188],"but":[189],"minimal":[191],"expensive":[192,218],"human":[193],"feedback.":[194],"demonstrate":[196],"our":[197],"through":[199],"experiments":[200],"simulation":[202],"on":[204],"hardware,":[205],"showing":[206],"UPS":[208],"disentangle":[210],"confident,":[211],"ambiguous,":[212],"scenarios":[215],"minimizes":[217],"user":[219],"compared":[221],"uncalibrated":[223],"baselines":[224],"prior":[226],"human-":[227],"robot-gated":[229],"continual":[230],"approaches.":[232],"Videos":[233],"be":[235],"found":[236],"https://jessie-yuan.github.io/ups/":[238]},"counts_by_year":[],"updated_date":"2026-04-04T16:13:02.066488","created_date":"2026-02-28T00:00:00"}