{"id":"https://openalex.org/W7147504653","doi":"https://doi.org/10.48550/arxiv.2603.27281","title":"HiFlow: Tokenization-Free Scale-Wise Autoregressive Policy Learning via Flow Matching","display_name":"HiFlow: Tokenization-Free Scale-Wise Autoregressive Policy Learning via Flow Matching","publication_year":2026,"publication_date":"2026-03-28","ids":{"openalex":"https://openalex.org/W7147504653","doi":"https://doi.org/10.48550/arxiv.2603.27281"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.27281","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.27281","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":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.2603.27281","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5115637817","display_name":"Daichi Yashima","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yashima, Daichi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5132644025","display_name":"Koki Seno","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Seno, Koki","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5073732915","display_name":"Shuhei Kurita","orcid":"https://orcid.org/0000-0001-7415-3120"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kurita, Shuhei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5075065777","display_name":"Yusuke Oda","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Oda, Yusuke","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5033744547","display_name":"Komei Sugiura","orcid":"https://orcid.org/0000-0002-0261-0510"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sugiura, Komei","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/T10462","display_name":"Reinforcement Learning in Robotics","score":0.362199991941452,"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":0.362199991941452,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.20499999821186066,"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/T10653","display_name":"Robot Manipulation and Learning","score":0.0982000008225441,"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/autoregressive-model","display_name":"Autoregressive model","score":0.8075000047683716},{"id":"https://openalex.org/keywords/codebook","display_name":"Codebook","score":0.47600001096725464},{"id":"https://openalex.org/keywords/action","display_name":"Action (physics)","score":0.4153999984264374},{"id":"https://openalex.org/keywords/leverage","display_name":"Leverage (statistics)","score":0.37779998779296875},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.36970001459121704},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.32690000534057617},{"id":"https://openalex.org/keywords/star-model","display_name":"STAR model","score":0.32330000400543213},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.3158999979496002}],"concepts":[{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.8075000047683716},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7268999814987183},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5321999788284302},{"id":"https://openalex.org/C127759330","wikidata":"https://www.wikidata.org/wiki/Q637416","display_name":"Codebook","level":2,"score":0.47600001096725464},{"id":"https://openalex.org/C2780791683","wikidata":"https://www.wikidata.org/wiki/Q846785","display_name":"Action (physics)","level":2,"score":0.4153999984264374},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.37779998779296875},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.36970001459121704},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3434000015258789},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.32690000534057617},{"id":"https://openalex.org/C194657046","wikidata":"https://www.wikidata.org/wiki/Q7394685","display_name":"STAR model","level":4,"score":0.32330000400543213},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.3158999979496002},{"id":"https://openalex.org/C42536954","wikidata":"https://www.wikidata.org/wiki/Q7049462","display_name":"Nonlinear autoregressive exogenous model","level":3,"score":0.30959999561309814},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.303600013256073},{"id":"https://openalex.org/C165064840","wikidata":"https://www.wikidata.org/wiki/Q1321061","display_name":"Matching (statistics)","level":2,"score":0.29580000042915344},{"id":"https://openalex.org/C160633673","wikidata":"https://www.wikidata.org/wiki/Q355198","display_name":"Pixel","level":2,"score":0.2849000096321106},{"id":"https://openalex.org/C2779803651","wikidata":"https://www.wikidata.org/wiki/Q5282088","display_name":"Discriminator","level":3,"score":0.28189998865127563},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.27639999985694885},{"id":"https://openalex.org/C2781181686","wikidata":"https://www.wikidata.org/wiki/Q4226068","display_name":"Coherence (philosophical gambling strategy)","level":2,"score":0.27549999952316284},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.2703999876976013},{"id":"https://openalex.org/C137703641","wikidata":"https://www.wikidata.org/wiki/Q4874480","display_name":"Bayesian vector autoregression","level":3,"score":0.2623000144958496},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.25290000438690186}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.27281","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.27281","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":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.2603.27281","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.27281","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":null,"license_id":null,"version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Preprint"},"sustainable_development_goals":[{"display_name":"Peace, Justice and strong institutions","score":0.4644832909107208,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Coarse-to-fine":[0],"autoregressive":[1,17,93,169],"modeling":[2],"has":[3],"recently":[4],"shown":[5],"strong":[6],"promise":[7],"for":[8,54,68,146],"visuomotor":[9],"policy":[10,94],"learning,":[11],"combining":[12],"the":[13,20,144],"inference":[14],"efficiency":[15],"of":[16,24],"methods":[18,164],"with":[19],"global":[21],"trajectory":[22],"coherence":[23],"diffusion-based":[25,166],"policies.":[26,170],"However,":[27],"existing":[28,163],"approaches":[29],"rely":[30],"on":[31,98,151],"discrete":[32],"action":[33,38,106,110,120],"tokenizers":[34],"that":[35,60,95,126,159],"map":[36],"continuous":[37,66,100,105],"sequences":[39],"to":[40,122],"codebook":[41],"indices,":[42],"a":[43,77,90,140,147],"design":[44],"inherited":[45],"from":[46,108],"image":[47],"generation":[48],"where":[49],"learned":[50],"compression":[51],"is":[52,136],"necessary":[53],"high-dimensional":[55],"pixel":[56],"data.":[57],"We":[58],"observe":[59],"robot":[61],"actions":[62],"are":[63,127],"inherently":[64],"low-dimensional":[65],"vectors,":[67],"which":[69],"such":[70],"tokenization":[71],"introduces":[72],"unnecessary":[73],"quantization":[74],"error":[75],"and":[76,155,167],"multi-stage":[78],"training":[79],"pipeline.":[80],"In":[81],"this":[82],"work,":[83],"we":[84],"propose":[85],"Hierarchical":[86],"Flow":[87],"Policy":[88],"(HiFlow),":[89],"tokenization-free":[91],"coarse-to-fine":[92],"operates":[96],"directly":[97],"raw":[99],"actions.":[101],"HiFlow":[102,160],"constructs":[103],"multi-scale":[104],"targets":[107],"each":[109],"chunk":[111],"via":[112],"simple":[113],"temporal":[114,131],"pooling.":[115],"Specifically,":[116],"it":[117],"averages":[118],"contiguous":[119],"windows":[121],"produce":[123],"coarse":[124],"summaries":[125],"refined":[128],"at":[129],"finer":[130],"resolutions.":[132],"The":[133],"entire":[134],"model":[135],"trained":[137],"end-to-end":[138],"in":[139],"single":[141],"stage,":[142],"eliminating":[143],"need":[145],"separate":[148],"tokenizer.":[149],"Experiments":[150],"MimicGen,":[152],"RoboTwin":[153],"2.0,":[154],"real-world":[156],"environments":[157],"demonstrate":[158],"consistently":[161],"outperforms":[162],"including":[165],"tokenization-based":[168]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-04-02T00:00:00"}