{"id":"https://openalex.org/W7165866143","doi":"https://doi.org/10.48550/arxiv.2606.26006","title":"FORCE: Efficient VLA Reinforcement Fine-Tuning via Value-Calibrated Warm-up and Self-Distillation","display_name":"FORCE: Efficient VLA Reinforcement Fine-Tuning via Value-Calibrated Warm-up and Self-Distillation","publication_year":2026,"publication_date":"2026-06-24","ids":{"openalex":"https://openalex.org/W7165866143","doi":"https://doi.org/10.48550/arxiv.2606.26006"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.26006","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.26006","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2606.26006","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139380515","display_name":"Shuyi Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Shuyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139349623","display_name":"Yunfan Lou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lou, Yunfan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5059304846","display_name":"Hongyang Cheng","orcid":"https://orcid.org/0000-0001-7652-8600"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cheng, Hongyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101893660","display_name":"Yi Guo","orcid":"https://orcid.org/0009-0006-1890-6726"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Guo, Yichen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5134269825","display_name":"Chuyao Fu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fu, Chuyao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139338951","display_name":"Yaoxu Lyu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lyu, Yaoxu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139369344","display_name":"Xiaojie Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Xiaojie","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139387335","display_name":"Haoran Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Haoran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139359324","display_name":"Pengwei Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Pengwei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139353647","display_name":"Zhongyuan Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Zhongyuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139312256","display_name":"Shanghang Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Shanghang","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":0.4681999981403351,"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":0.4681999981403351,"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":0.2535000145435333,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.07829999923706055,"subfield":{"id":"https://openalex.org/subfields/2207","display_name":"Control and Systems 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/reinforcement-learning","display_name":"Reinforcement learning","score":0.7366999983787537},{"id":"https://openalex.org/keywords/ceiling","display_name":"Ceiling (cloud)","score":0.4546999931335449},{"id":"https://openalex.org/keywords/forcing","display_name":"Forcing (mathematics)","score":0.43700000643730164},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.37619999051094055},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.35249999165534973},{"id":"https://openalex.org/keywords/filter","display_name":"Filter (signal processing)","score":0.35249999165534973},{"id":"https://openalex.org/keywords/core","display_name":"Core (optical fiber)","score":0.34850001335144043}],"concepts":[{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.7366999983787537},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6933000087738037},{"id":"https://openalex.org/C2777489069","wikidata":"https://www.wikidata.org/wiki/Q1589822","display_name":"Ceiling (cloud)","level":2,"score":0.4546999931335449},{"id":"https://openalex.org/C197115733","wikidata":"https://www.wikidata.org/wiki/Q1003136","display_name":"Forcing (mathematics)","level":2,"score":0.43700000643730164},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.37619999051094055},{"id":"https://openalex.org/C106131492","wikidata":"https://www.wikidata.org/wiki/Q3072260","display_name":"Filter (signal processing)","level":2,"score":0.35249999165534973},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.35249999165534973},{"id":"https://openalex.org/C2164484","wikidata":"https://www.wikidata.org/wiki/Q5170150","display_name":"Core (optical fiber)","level":2,"score":0.34850001335144043},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.33719998598098755},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.32429999113082886},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.3239000141620636},{"id":"https://openalex.org/C112972136","wikidata":"https://www.wikidata.org/wiki/Q7595718","display_name":"Stability (learning theory)","level":2,"score":0.3221000134944916},{"id":"https://openalex.org/C67203356","wikidata":"https://www.wikidata.org/wiki/Q1321905","display_name":"Reinforcement","level":2,"score":0.2962999939918518},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.2888999879360199},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.27790001034736633},{"id":"https://openalex.org/C177264268","wikidata":"https://www.wikidata.org/wiki/Q1514741","display_name":"Set (abstract data type)","level":2,"score":0.25999999046325684},{"id":"https://openalex.org/C13687954","wikidata":"https://www.wikidata.org/wiki/Q4826847","display_name":"Autonomous agent","level":2,"score":0.2500999867916107}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.26006","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.26006","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2606.26006","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.26006","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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-Action":[0],"(VLA)":[1],"models":[2],"are":[3,119],"often":[4,54],"constrained":[5],"by":[6,11,50,71,153,158],"the":[7,87,91,95,107,122,135,163],"imitation":[8],"ceiling":[9],"imposed":[10],"sub-optimal":[12],"data.":[13],"While":[14],"Reinforcement":[15],"Learning":[16],"(RL)":[17],"fine-tuning":[18,70],"can":[19],"surpass":[20],"this":[21,98,170],"limit,":[22],"it":[23,161],"is":[24],"notoriously":[25],"sample":[26],"inefficient.":[27],"This":[28],"challenge":[29],"arises":[30],"from":[31],"two":[32],"core":[33],"issues:":[34],"(1)":[35],"catastrophic":[36],"initial":[37],"unlearning":[38],"due":[39],"to":[40,85],"an":[41],"unstable":[42],"Q-function":[43,100],"and":[44,112,131,134,148,168,183],"(2)":[45],"inefficient":[46],"policy":[47,123],"updates":[48],"caused":[49],"low-quality":[51],"exploration":[52],"data,":[53,114],"forcing":[55],"a":[56,65,78,103,141,177],"reliance":[57],"on":[58,128],"costly":[59],"human":[60,174],"interventions.":[61],"We":[62,125],"introduce":[63],"FORCE,":[64],"3-stage":[66],"framework":[67],"that":[68,138],"stabilizes":[69],"tackling":[72],"both":[73,106],"issues.":[74],"FORCE":[75,127,139],"first":[76],"incorporates":[77],"Value-Calibrated":[79],"Warm-Up":[80],"phase,":[81],"utilizing":[82],"on-policy":[83],"rollouts":[84],"mitigate":[86],"distributional":[88],"shift":[89],"of":[90],"Q-function.":[92],"Subsequently,":[93],"during":[94],"online":[96],"stage,":[97],"calibrated":[99],"acts":[101],"as":[102],"filter":[104],"for":[105,121],"policy's":[108],"own":[109],"action":[110],"proposals":[111],"expert":[113],"ensuring":[115],"only":[116],"high-value":[117],"actions":[118],"used":[120],"update.":[124],"evaluate":[126],"various":[129],"simulation":[130],"real-world":[132],"tasks,":[133],"result":[136],"shows":[137],"achieves":[140,169],"79%":[142],"absolute":[143],"improvement":[144],"in":[145],"success":[146,165],"rates":[147],"outperform":[149],"prior":[150],"RL":[151],"methods":[152],"10%,":[154],"while":[155],"accelerating":[156],"training":[157],"32.5%.":[159],"Critically,":[160],"mitigates":[162],"common":[164],"rate":[166],"drop":[167],"robust":[171],"performance":[172],"without":[173],"intervention,":[175],"marking":[176],"significant":[178],"step":[179],"towards":[180],"deploying":[181],"capable":[182],"autonomous":[184],"robotic":[185],"agents.":[186]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-06-26T00:00:00"}