{"id":"https://openalex.org/W7160865889","doi":"https://doi.org/10.48550/arxiv.2605.07560","title":"How to Utilize Failure Demo Data?: Effective Data Selection for Imitation Learning Using Distribution Differences in Attention Mechanism","display_name":"How to Utilize Failure Demo Data?: Effective Data Selection for Imitation Learning Using Distribution Differences in Attention Mechanism","publication_year":2026,"publication_date":"2026-05-08","ids":{"openalex":"https://openalex.org/W7160865889","doi":"https://doi.org/10.48550/arxiv.2605.07560"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.07560","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.07560","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":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.2605.07560","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5127971367","display_name":"Kana Miyamoto","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Miyamoto, Kana","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042849965","display_name":"Kanata Suzuki","orcid":"https://orcid.org/0000-0001-7122-7649"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Suzuki, Kanata","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5135902716","display_name":"Tetsuya Ogata","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ogata, Tetsuya","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/T10653","display_name":"Robot Manipulation and Learning","score":0.4505999982357025,"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":0.4505999982357025,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.1111999973654747,"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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.08229999989271164,"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/task","display_name":"Task (project management)","score":0.5672000050544739},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.5511000156402588},{"id":"https://openalex.org/keywords/data-collection","display_name":"Data collection","score":0.4323999881744385},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.4194999933242798},{"id":"https://openalex.org/keywords/imitation","display_name":"Imitation","score":0.4104999899864197},{"id":"https://openalex.org/keywords/action-selection","display_name":"Action selection","score":0.3944999873638153},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.35580000281333923},{"id":"https://openalex.org/keywords/scheme","display_name":"Scheme (mathematics)","score":0.3361999988555908}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7465000152587891},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5864999890327454},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.569599986076355},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.5672000050544739},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.5511000156402588},{"id":"https://openalex.org/C133462117","wikidata":"https://www.wikidata.org/wiki/Q4929239","display_name":"Data collection","level":2,"score":0.4323999881744385},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.4194999933242798},{"id":"https://openalex.org/C126388530","wikidata":"https://www.wikidata.org/wiki/Q1131737","display_name":"Imitation","level":2,"score":0.4104999899864197},{"id":"https://openalex.org/C166109690","wikidata":"https://www.wikidata.org/wiki/Q4677422","display_name":"Action selection","level":3,"score":0.3944999873638153},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.35580000281333923},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.3361999988555908},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.3188000023365021},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.30970001220703125},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.3061999976634979},{"id":"https://openalex.org/C2780440489","wikidata":"https://www.wikidata.org/wiki/Q5227278","display_name":"Data-driven","level":2,"score":0.2946000099182129},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.28929999470710754},{"id":"https://openalex.org/C2776145971","wikidata":"https://www.wikidata.org/wiki/Q30673951","display_name":"Labeled data","level":2,"score":0.2879999876022339},{"id":"https://openalex.org/C163164238","wikidata":"https://www.wikidata.org/wiki/Q2737027","display_name":"Failure rate","level":2,"score":0.28189998865127563},{"id":"https://openalex.org/C48677424","wikidata":"https://www.wikidata.org/wiki/Q6888088","display_name":"Mode (computer interface)","level":2,"score":0.2815000116825104},{"id":"https://openalex.org/C66283442","wikidata":"https://www.wikidata.org/wiki/Q1389268","display_name":"Failure mode and effects analysis","level":2,"score":0.2736999988555908},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.27239999175071716},{"id":"https://openalex.org/C51632099","wikidata":"https://www.wikidata.org/wiki/Q3985153","display_name":"Training set","level":2,"score":0.25440001487731934}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.07560","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.07560","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":"Preprint"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2605.07560","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.07560","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":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":{"Imitation":[0],"learning":[1,142],"for":[2,26,141],"robotic":[3,164],"tasks":[4],"has":[5],"relied":[6],"primarily":[7],"on":[8,12],"policies":[9],"trained":[10,126],"only":[11],"successful":[13,108,146],"demonstrations,":[14],"although":[15],"failures":[16],"are":[17,139],"unavoidable":[18],"during":[19,52],"human":[20],"data":[21,29,32,50,53,129,165],"collection.":[22,54],"Many":[23],"existing":[24],"approaches":[25],"exploiting":[27],"failure":[28,49,105,112,128,136],"require":[30],"additional":[31],"processing":[33],"or":[34],"iterative":[35],"policy":[36],"updates":[37],"through":[38],"autonomous":[39],"rollouts,":[40],"making":[41],"it":[42],"difficult":[43],"to":[44,88,110],"directly":[45],"and":[46,69,107,130],"stably":[47],"utilize":[48],"accumulated":[51],"In":[55],"this":[56],"work,":[57],"we":[58,93],"propose":[59],"a":[60,95],"method":[61,120,154],"that":[62,98,117,131,138,151],"learns":[63],"latent":[64,80],"representations":[65],"of":[66,160],"success-failure":[67],"discrepancies":[68],"incorporates":[70],"them":[71],"into":[72],"the":[73,85,100,118,132,152],"attention":[74,101],"mechanism.":[75],"During":[76],"inference,":[77],"an":[78],"appropriate":[79],"mode":[81],"is":[82],"selected":[83],"from":[84],"initial":[86],"observation":[87],"improve":[89],"action":[90],"stability.":[91],"Furthermore,":[92],"introduce":[94],"post-training":[96],"metric":[97,134],"quantifies":[99],"discrepancy":[102],"between":[103],"each":[104],"sample":[106],"demonstrations":[109,162],"select":[111],"data.":[113],"Simulation":[114],"results":[115,149],"show":[116],"proposed":[119,133,153],"improves":[121],"task":[122],"success":[123],"rates":[124],"when":[125,143],"with":[127,145],"identifies":[135],"samples":[137],"beneficial":[140],"combined":[144],"demonstrations.":[147],"These":[148],"suggest":[150],"can":[155],"support":[156],"more":[157],"efficient":[158],"use":[159],"collected":[161],"in":[163],"collection":[166],"pipelines.":[167]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-12T00:00:00"}