{"id":"https://openalex.org/W7204232380","doi":"https://doi.org/10.48550/arxiv.2608.24042","title":"Hierarchical Skill Retrieval for Data-Efficient Adaptation of Vision-Language-Action Models","display_name":"Hierarchical Skill Retrieval for Data-Efficient Adaptation of Vision-Language-Action Models","publication_year":2026,"publication_date":"2026-08-25","ids":{"openalex":"https://openalex.org/W7204232380","doi":"https://doi.org/10.48550/arxiv.2608.24042"},"language":"en","primary_location":{"id":"pmh:oai:arXiv.org:2608.24042","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24042","pdf_url":"https://arxiv.org/pdf/2608.24042","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"volume":null,"type":"preprint","indexed_in":["arxiv","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2608.24042","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114337935","display_name":"Haoran Hao","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Hao, Haoran","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041669967","display_name":"Shahram Najam Syed","orcid":"https://orcid.org/0009-0007-5752-0786","observed_orcids":["https://orcid.org/0009-0007-5752-0786"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Syed, Shahram Najam","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5152486128","display_name":"Jeff Schneider","orcid":null,"observed_orcids":[]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Schneider, Jeff","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5005587345","display_name":"Jeffrey Ichnowski","orcid":"https://orcid.org/0000-0003-4874-9478","observed_orcids":["https://orcid.org/0000-0003-4874-9478"]},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ichnowski, Jeffrey","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":true,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.9911999702453613,"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"}},"topics":[{"id":"https://openalex.org/T10653","display_name":"Robot Manipulation and Learning","score":0.9911999702453613,"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"}},{"id":"https://openalex.org/T10462","display_name":"Reinforcement Learning in Robotics","score":0.007000000216066837,"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":0.00039999998989515007,"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/vision-language-action-models","display_name":"vision-language-action models","score":0.9120000004768372},{"id":"https://openalex.org/keywords/robotic-manipulation","display_name":"robotic manipulation","score":0.8889999985694885},{"id":"https://openalex.org/keywords/skill-decomposition","display_name":"skill decomposition","score":0.578000009059906},{"id":"https://openalex.org/keywords/libero-benchmark","display_name":"LIBERO benchmark","score":0.8870000243186951}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8079000115394592},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.6258000135421753},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.6215999722480774},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.603600025177002},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5792999863624573},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.48170000314712524},{"id":"https://openalex.org/C43214815","wikidata":"https://www.wikidata.org/wiki/Q7310987","display_name":"Reliability (semiconductor)","level":3,"score":0.48019999265670776},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.4763999879360199},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.4142000079154968},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3959999978542328},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.362199991941452},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.35530000925064087},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.34599998593330383},{"id":"https://openalex.org/C144986985","wikidata":"https://www.wikidata.org/wiki/Q871236","display_name":"Hierarchical database model","level":2,"score":0.335099995136261},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.31859999895095825},{"id":"https://openalex.org/C2776505523","wikidata":"https://www.wikidata.org/wiki/Q4785468","display_name":"Plan (archaeology)","level":2,"score":0.3176000118255615},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3158999979496002},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3147999942302704},{"id":"https://openalex.org/C124681953","wikidata":"https://www.wikidata.org/wiki/Q339062","display_name":"Decomposition","level":2,"score":0.31470000743865967},{"id":"https://openalex.org/C2778583558","wikidata":"https://www.wikidata.org/wiki/Q771245","display_name":"Code reuse","level":3,"score":0.2964000105857849},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.2912999987602234},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.2854999899864197},{"id":"https://openalex.org/C132758656","wikidata":"https://www.wikidata.org/wiki/Q5307365","display_name":"Dreyfus model of skill acquisition","level":2,"score":0.273499995470047},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.2700999975204468},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2653999924659729},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.2653999924659729},{"id":"https://openalex.org/C2775955345","wikidata":"https://www.wikidata.org/wiki/Q7449071","display_name":"Semantic mapping","level":2,"score":0.2524999976158142}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:oai:arXiv.org:2608.24042","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24042","pdf_url":"https://arxiv.org/pdf/2608.24042","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2608.24042","doi":"https://doi.org/10.48550/arxiv.2608.24042","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2608.24042","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,"listed_in":[],"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":false,"raw_source_name":null,"raw_type":"Preprint"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2608.24042","doi":null,"is_oa":true,"landing_page_url":"https://arxiv.org/abs/2608.24042","pdf_url":"https://arxiv.org/pdf/2608.24042","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,"listed_in":[],"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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"study_designs":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7204232380.pdf","grobid_xml":"https://content.openalex.org/works/W7204232380.grobid-xml"},"referenced_works_count":0,"referenced_works":[],"related_works":["https://openalex.org/W4415955379","https://openalex.org/W4416157228","https://openalex.org/W7215050923","https://openalex.org/W7220712189","https://openalex.org/W7204458710","https://openalex.org/W7164881028","https://openalex.org/W7162331955","https://openalex.org/W4414757762","https://openalex.org/W7160948833","https://openalex.org/W7169576848"],"abstract_inverted_index":{"While":[0],"Vision-Language-Action":[1],"(VLA)":[2],"models":[3],"pretrained":[4],"on":[5,46,111,175],"large-scale":[6],"robot":[7,14,182],"datasets":[8],"provide":[9],"a":[10,31,88,99,160],"strong":[11],"foundation":[12],"for":[13,38,91,129,211],"manipulation,":[15],"their":[16],"performance":[17],"can":[18],"degrade":[19],"when":[20],"adapted":[21],"to":[22,34,140],"new":[23],"tasks":[24,184],"with":[25,137,150],"limited":[26],"task-specific":[27,172],"demonstrations.":[28],"Retrieval":[29,86],"offers":[30],"practical":[32],"way":[33],"reuse":[35],"existing":[36,42],"demonstrations":[37,142],"data-efficient":[39,92,212],"adaptation,":[40],"but":[41,72],"methods":[43],"often":[44,76],"rely":[45],"visual":[47],"similarity,":[48],"state-action":[49],"representations,":[50],"or":[51],"task-level":[52],"language":[53,135],"matching.":[54],"These":[55,202],"approaches":[56],"may":[57],"overlook":[58],"the":[59,120,151,157,176,189,198,205],"hierarchical":[60],"structure":[61],"of":[62,207],"long-horizon":[63],"manipulation":[64,183],"tasks,":[65],"where":[66],"complete":[67],"task":[68,101],"matches":[69],"are":[70,75,144,218],"rare":[71],"reusable":[73],"skills":[74],"abundant.":[77],"To":[78],"address":[79],"this":[80],"challenge,":[81],"we":[82,155],"propose":[83],"Hierarchical":[84],"Skill":[85],"(HSR),":[87],"retrieval":[89,136,210],"framework":[90],"VLA":[93,213],"adaptation.":[94,173,214],"Specifically,":[95],"HSR":[96,187],"first":[97],"decomposes":[98],"target":[100,152],"into":[102],"candidate":[103],"skill":[104,116,169],"sequences.":[105],"It":[106],"evaluates":[107],"each":[108],"plan":[109],"based":[110],"both":[112,145],"semantic":[113],"plausibility":[114],"and":[115,148,163,179,195,216],"reliability":[117],"estimated":[118],"from":[119,171],"prior":[121],"dataset.":[122],"The":[123],"selected":[124],"decomposition":[125],"is":[126],"then":[127],"used":[128],"hybrid":[130],"retrieval.":[131],"This":[132],"combines":[133],"subtask-level":[134],"behavior-feature":[138],"reranking":[139],"identify":[141],"that":[143,186],"semantically":[146],"relevant":[147],"compatible":[149],"task.":[153],"Finally,":[154],"adapt":[156],"policy":[158],"through":[159],"two-stage":[161],"pretraining":[162],"finetuning":[164],"pipeline,":[165],"which":[166],"separates":[167],"general":[168],"acquisition":[170],"Experiments":[174],"LIBERO":[177],"benchmark":[178],"several":[180],"real-world":[181],"show":[185],"improves":[188],"average":[190],"success":[191],"rate":[192],"by":[193],"10.3%":[194],"21.3%":[196],"over":[197],"strongest":[199],"baseline,":[200],"respectively.":[201],"results":[203],"demonstrate":[204],"effectiveness":[206],"structured":[208],"skill-level":[209],"Videos":[215],"code":[217],"available":[219],"at":[220],"https://hoar012.github.io/HSR-Project.":[221]},"counts_by_year":[],"updated_date":"2026-10-28T10:33:24.094632","created_date":"2026-08-27T00:00:00"}