{"id":"https://openalex.org/W7138971171","doi":"https://doi.org/10.48550/arxiv.2603.17573","title":"HeiSD: Hybrid Speculative Decoding for Embodied Vision-Language-Action Models with Kinematic Awareness","display_name":"HeiSD: Hybrid Speculative Decoding for Embodied Vision-Language-Action Models with Kinematic Awareness","publication_year":2026,"publication_date":"2026-03-18","ids":{"openalex":"https://openalex.org/W7138971171","doi":"https://doi.org/10.48550/arxiv.2603.17573"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.17573","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17573","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"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":"https://doi.org/10.48550/arxiv.2603.17573","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129943196","display_name":"Zihao Zheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zheng, Zihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129923496","display_name":"Zhihao Mao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mao, Zhihao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5104294481","display_name":"Sicheng Tian","orcid":"https://orcid.org/0000-0002-7134-9774"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Tian, Sicheng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129919962","display_name":"Maoliang Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Maoliang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129930807","display_name":"Jiayu Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Jiayu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130144985","display_name":"Xinhao Sun","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Sun, Xinhao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127941642","display_name":"Zhaobo Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Zhaobo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129861458","display_name":"Xuanzhe Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Xuanzhe","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127879585","display_name":"Donggang Cao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cao, Donggang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130203989","display_name":"Hong Mei","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Mei, Hong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5129992510","display_name":"Xiang Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Xiang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":11,"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":0.6010000109672546,"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":0.6010000109672546,"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":0.09939999878406525,"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":0.03290000185370445,"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/decoding-methods","display_name":"Decoding methods","score":0.6323000192642212},{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.5997999906539917},{"id":"https://openalex.org/keywords/metric","display_name":"Metric (unit)","score":0.5838000178337097},{"id":"https://openalex.org/keywords/kinematics","display_name":"Kinematics","score":0.5705000162124634},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.5672000050544739},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5385000109672546},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5203999876976013},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.5077000260353088}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7487000226974487},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.6323000192642212},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.5997999906539917},{"id":"https://openalex.org/C176217482","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Metric (unit)","level":2,"score":0.5838000178337097},{"id":"https://openalex.org/C39920418","wikidata":"https://www.wikidata.org/wiki/Q11476","display_name":"Kinematics","level":2,"score":0.5705000162124634},{"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/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5385000109672546},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5203999876976013},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5182999968528748},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.5077000260353088},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.48339998722076416},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.4560999870300293},{"id":"https://openalex.org/C50897621","wikidata":"https://www.wikidata.org/wiki/Q2665508","display_name":"Hybrid system","level":2,"score":0.43869999051094055},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.3483000099658966},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3379000127315521},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.33390000462532043},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3257000148296356},{"id":"https://openalex.org/C12713177","wikidata":"https://www.wikidata.org/wiki/Q1900281","display_name":"Perspective (graphical)","level":2,"score":0.32260000705718994},{"id":"https://openalex.org/C147297375","wikidata":"https://www.wikidata.org/wiki/Q6674930","display_name":"Look-ahead","level":2,"score":0.2865999937057495},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.27869999408721924},{"id":"https://openalex.org/C173246807","wikidata":"https://www.wikidata.org/wiki/Q7833062","display_name":"Trajectory optimization","level":3,"score":0.2605000138282776},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.2590999901294708},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.2565999925136566},{"id":"https://openalex.org/C2780898871","wikidata":"https://www.wikidata.org/wiki/Q860554","display_name":"Performance metric","level":2,"score":0.25029999017715454}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.17573","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17573","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":null,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.17573","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.17573","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":null,"is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},"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],"have":[3],"become":[4],"the":[5,39,54,81,91,117,125,160],"mainstream":[6],"solution":[7],"for":[8],"robot":[9],"control,":[10],"but":[11],"suffer":[12],"from":[13],"slow":[14],"inference":[15],"speeds.":[16],"Speculative":[17],"Decoding":[18],"(SD)":[19],"is":[20],"a":[21,57,74,130,139,143,151,169,185],"promising":[22],"acceleration":[23],"method":[24,134],"which":[25,137],"can":[26],"be":[27],"divided":[28],"into":[29],"two":[30,40,62],"categories:":[31],"drafter-based":[32],"SD":[33,97,132],"and":[34,45,83,107,142,178],"retrieval-based":[35,111,131],"SD.":[36],"Each":[37],"of":[38,76,85,171],"methods":[41,63],"demonstrates":[42],"complementary":[43],"advantages":[44,82],"limitations":[46],"when":[47],"applied":[48],"to":[49,53,79,157,173],"VLA":[50,99],"models,":[51],"leading":[52],"hypothesis":[55],"that":[56,166],"hybrid":[58,86,96,118,161],"approach":[59],"integrating":[60],"these":[61],"will":[64],"yield":[65],"better":[66],"performance.":[67],"In":[68],"this":[69],"paper,":[70],"we":[71,123,149],"first":[72],"conduct":[73],"series":[75],"detailed":[77],"analyses":[78],"reveal":[80],"feasibility":[84],"utilization.":[87],"However,":[88],"even":[89],"with":[90],"aforementioned":[92],"key":[93],"insights,":[94],"implementing":[95],"in":[98,110,115,135,155,175,180],"models":[100],"presents":[101],"several":[102],"challenges:":[103],"(1)":[104],"draft":[105],"rejection":[106],"persistent":[108],"errors":[109],"SD;":[112],"(2)":[113],"difficulty":[114],"determining":[116],"boundary.":[119,162],"To":[120],"address":[121],"these,":[122],"propose":[124,129],"HeiSD":[126,156,167],"framework.":[127],"We":[128],"optimization":[133],"HeiSD,":[136],"contains":[138],"verify-skip":[140],"mechanism":[141],"sequence-wise":[144],"relaxed":[145],"acceptance":[146],"strategy.":[147],"Moreover,":[148],"proposed":[150],"kinematic-based":[152],"fused":[153],"metric":[154],"automatically":[158],"determine":[159],"Experimental":[163],"results":[164],"demonstrate":[165],"attains":[168],"speedup":[170],"up":[172],"2.45x":[174],"simulation":[176],"benchmarks":[177],"2.06x~2.41x":[179],"real-world":[181],"scenarios,":[182],"while":[183],"sustaining":[184],"high":[186],"task":[187],"success":[188],"rate.":[189]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-20T00:00:00"}