{"id":"https://openalex.org/W7131862339","doi":"https://doi.org/10.48550/arxiv.2602.22896","title":"DySL-VLA: Efficient Vision-Language-Action Model Inference via Dynamic-Static Layer-Skipping for Robot Manipulation","display_name":"DySL-VLA: Efficient Vision-Language-Action Model Inference via Dynamic-Static Layer-Skipping for Robot Manipulation","publication_year":2026,"publication_date":"2026-02-26","ids":{"openalex":"https://openalex.org/W7131862339","doi":"https://doi.org/10.48550/arxiv.2602.22896"},"language":null,"primary_location":{"id":"pmh:doi:10.48550/arxiv.2602.22896","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"type":"article","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":null,"any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5114950458","display_name":"Zebin Yang","orcid":"https://orcid.org/0009-0007-0672-6389"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Zebin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121216079","display_name":"Yijiahao Qi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Qi, Yijiahao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5121365515","display_name":"Tong Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Tong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127053022","display_name":"Bo Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Bo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5127181508","display_name":"Shaoshan Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Shaoshan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5127074694","display_name":"Meng Li","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Meng","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":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.25355921,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"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":0.8561000227928162,"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":0.8561000227928162,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.03150000050663948,"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":0.02199999988079071,"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/speedup","display_name":"Speedup","score":0.6610999703407288},{"id":"https://openalex.org/keywords/task","display_name":"Task (project management)","score":0.6342999935150146},{"id":"https://openalex.org/keywords/obstacle","display_name":"Obstacle","score":0.6166999936103821},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.6150000095367432},{"id":"https://openalex.org/keywords/code","display_name":"Code (set theory)","score":0.5584999918937683},{"id":"https://openalex.org/keywords/robot","display_name":"Robot","score":0.5239999890327454},{"id":"https://openalex.org/keywords/baseline","display_name":"Baseline (sea)","score":0.4122999906539917},{"id":"https://openalex.org/keywords/factor","display_name":"Factor (programming language)","score":0.38359999656677246},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.35839998722076416}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7947999835014343},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6610999703407288},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.6342999935150146},{"id":"https://openalex.org/C2776650193","wikidata":"https://www.wikidata.org/wiki/Q264661","display_name":"Obstacle","level":2,"score":0.6166999936103821},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6150000095367432},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.593999981880188},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.5584999918937683},{"id":"https://openalex.org/C90509273","wikidata":"https://www.wikidata.org/wiki/Q11012","display_name":"Robot","level":2,"score":0.5239999890327454},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.42239999771118164},{"id":"https://openalex.org/C12725497","wikidata":"https://www.wikidata.org/wiki/Q810247","display_name":"Baseline (sea)","level":2,"score":0.4122999906539917},{"id":"https://openalex.org/C2781039887","wikidata":"https://www.wikidata.org/wiki/Q1391724","display_name":"Factor (programming language)","level":2,"score":0.38359999656677246},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.35839998722076416},{"id":"https://openalex.org/C89611455","wikidata":"https://www.wikidata.org/wiki/Q6804646","display_name":"Mechanism (biology)","level":2,"score":0.3292999863624573},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.32829999923706055},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.3280999958515167},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.30329999327659607},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3019999861717224},{"id":"https://openalex.org/C204030448","wikidata":"https://www.wikidata.org/wiki/Q101017","display_name":"Distillation","level":2,"score":0.2957000136375427},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2831000089645386},{"id":"https://openalex.org/C43126263","wikidata":"https://www.wikidata.org/wiki/Q128751","display_name":"Source code","level":2,"score":0.2770000100135803},{"id":"https://openalex.org/C6683253","wikidata":"https://www.wikidata.org/wiki/Q7075535","display_name":"Obstacle avoidance","level":4,"score":0.2687999904155731},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.26420000195503235},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.259799987077713},{"id":"https://openalex.org/C60692881","wikidata":"https://www.wikidata.org/wiki/Q584529","display_name":"Humanoid robot","level":3,"score":0.257099986076355},{"id":"https://openalex.org/C34413123","wikidata":"https://www.wikidata.org/wiki/Q170978","display_name":"Robotics","level":3,"score":0.25679999589920044},{"id":"https://openalex.org/C5339829","wikidata":"https://www.wikidata.org/wiki/Q1425977","display_name":"Machine vision","level":2,"score":0.2567000091075897}],"mesh":[],"locations_count":2,"locations":[{"id":"pmh:doi:10.48550/arxiv.2602.22896","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},{"id":"doi:10.48550/arxiv.2602.22896","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2602.22896","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":"Preprint"}],"best_oa_location":{"id":"pmh:doi:10.48550/arxiv.2602.22896","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Vision-Language-Action":[0],"(VLA)":[1],"models":[2],"have":[3,48],"shown":[4],"remarkable":[5],"success":[6,157],"in":[7,156],"robotic":[8],"tasks":[9],"like":[10],"manipulation":[11],"by":[12,79,170],"fusing":[13],"a":[14,19,30,46,72,119,133,142,146,171,177],"language":[15],"model's":[16,21],"reasoning":[17],"with":[18],"vision":[20],"3D":[22],"understanding.":[23],"However,":[24],"their":[25],"high":[26,56],"computational":[27,77],"cost":[28,78],"remains":[29],"major":[31],"obstacle":[32],"for":[33],"real-world":[34],"applications":[35],"that":[36,42,75,151],"require":[37],"real-time":[38],"performance.":[39],"We":[40,130],"observe":[41],"the":[43,162,182],"actions":[44],"within":[45],"task":[47],"varying":[49],"levels":[50],"of":[51,173],"importance:":[52],"critical":[53],"steps":[54],"demand":[55],"precision,":[57],"while":[58,165],"less":[59],"important":[60],"ones":[61],"can":[62,106],"tolerate":[63],"more":[64],"variance.":[65],"Leveraging":[66],"this":[67],"insight,":[68],"we":[69,117],"propose":[70,132],"DySL-VLA,":[71],"novel":[73],"framework":[74],"addresses":[76],"dynamically":[80],"skipping":[81,121],"VLA":[82,144],"layers":[83,92,113],"based":[84],"on":[85,161,191],"each":[86],"action's":[87],"importance.":[88],"DySL-VLA":[89,152],"categorizes":[90],"its":[91],"into":[93,145],"two":[94],"types:":[95],"informative":[96],"layers,":[97,104],"which":[98,105],"are":[99],"consistently":[100],"executed,":[101],"and":[102,175],"incremental":[103],"be":[107],"selectively":[108],"skipped.":[109],"To":[110],"intelligently":[111],"skip":[112],"without":[114],"sacrificing":[115],"accuracy,":[116],"invent":[118],"prior-post":[120],"guidance":[122],"mechanism":[123],"to":[124,127,139,181],"determine":[125],"when":[126],"initiate":[128],"layer-skipping.":[129],"also":[131],"skip-aware":[134],"two-stage":[135],"knowledge":[136],"distillation":[137],"algorithm":[138],"efficiently":[140],"train":[141],"standard":[143],"DySL-VLA.":[147],"Our":[148,187],"experiments":[149],"indicate":[150],"achieves":[153],"2.1%":[154],"improvement":[155],"length":[158],"over":[159],"Deer-VLA":[160],"Calvin":[163],"dataset,":[164],"simultaneously":[166],"reducing":[167],"trainable":[168],"parameters":[169],"factor":[172],"85.7":[174],"providing":[176],"3.75x":[178],"speedup":[179],"relative":[180],"RoboFlamingo":[183],"baseline":[184],"at":[185],"iso-accuracy.":[186],"code":[188],"is":[189],"available":[190],"https://github.com/PKU-SEC-Lab/DYSL_VLA.":[192]},"counts_by_year":[],"updated_date":"2026-07-15T18:14:33.161393","created_date":"2026-02-28T00:00:00"}