{"id":"https://openalex.org/W7162330647","doi":"https://doi.org/10.48550/arxiv.2605.23163","title":"Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving","display_name":"Fast-dDrive: Efficient Block-Diffusion VLM for Autonomous Driving","publication_year":2026,"publication_date":"2026-05-22","ids":{"openalex":"https://openalex.org/W7162330647","doi":"https://doi.org/10.48550/arxiv.2605.23163"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2605.23163","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.23163","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.2605.23163","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5136923976","display_name":"Kewei Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Kewei","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136925395","display_name":"Jin Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Jin","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5114236320","display_name":"Sensen Gao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Gao, Sensen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136926606","display_name":"Chengyue Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Chengyue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136923094","display_name":"Yulong Cao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cao, Yulong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026925915","display_name":"Songyang Han","orcid":"https://orcid.org/0000-0001-8314-4832"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Songyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091869385","display_name":"Boris Ivanovic","orcid":"https://orcid.org/0000-0002-8698-202X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ivanovic, Boris","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5016851988","display_name":"Langechuan Liu","orcid":"https://orcid.org/0000-0001-5476-0169"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Langechuan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136982714","display_name":"Marco Pavone","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Pavone, Marco","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136944481","display_name":"Song Han","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Han, Song","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136969028","display_name":"Daquan Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Daquan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5136950292","display_name":"Enze Xie","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Enze","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/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.7706000208854675,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive 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/T11099","display_name":"Autonomous Vehicle Technology and Safety","score":0.7706000208854675,"subfield":{"id":"https://openalex.org/subfields/2203","display_name":"Automotive 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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.03700000047683716,"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":0.02889999933540821,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/trajectory","display_name":"Trajectory","score":0.6845999956130981},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.6291999816894531},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.5760999917984009},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5246000289916992},{"id":"https://openalex.org/keywords/scaling","display_name":"Scaling","score":0.460099995136261},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.42899999022483826},{"id":"https://openalex.org/keywords/decoding-methods","display_name":"Decoding methods","score":0.40630000829696655},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.39100000262260437},{"id":"https://openalex.org/keywords/multi-core-processor","display_name":"Multi-core processor","score":0.36890000104904175},{"id":"https://openalex.org/keywords/block","display_name":"Block (permutation group theory)","score":0.35910001397132874}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7383999824523926},{"id":"https://openalex.org/C13662910","wikidata":"https://www.wikidata.org/wiki/Q193139","display_name":"Trajectory","level":2,"score":0.6845999956130981},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.6291999816894531},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.5760999917984009},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5246000289916992},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.460099995136261},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.42899999022483826},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.40630000829696655},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.39100000262260437},{"id":"https://openalex.org/C78766204","wikidata":"https://www.wikidata.org/wiki/Q555032","display_name":"Multi-core processor","level":2,"score":0.36890000104904175},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3628999888896942},{"id":"https://openalex.org/C2777210771","wikidata":"https://www.wikidata.org/wiki/Q4927124","display_name":"Block (permutation group theory)","level":2,"score":0.35910001397132874},{"id":"https://openalex.org/C198082294","wikidata":"https://www.wikidata.org/wiki/Q3399648","display_name":"Position (finance)","level":2,"score":0.353300005197525},{"id":"https://openalex.org/C168167062","wikidata":"https://www.wikidata.org/wiki/Q1117970","display_name":"Component (thermodynamics)","level":2,"score":0.34880000352859497},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.33500000834465027},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3276999890804291},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.323199987411499},{"id":"https://openalex.org/C196083921","wikidata":"https://www.wikidata.org/wiki/Q7915758","display_name":"Variance (accounting)","level":2,"score":0.3192000091075897},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.3034000098705292},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C74172769","wikidata":"https://www.wikidata.org/wiki/Q1446839","display_name":"Routing (electronic design automation)","level":2,"score":0.3001999855041504},{"id":"https://openalex.org/C2776266606","wikidata":"https://www.wikidata.org/wiki/Q476482","display_name":"Test bench","level":2,"score":0.2948000133037567},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.28349998593330383},{"id":"https://openalex.org/C44154836","wikidata":"https://www.wikidata.org/wiki/Q45045","display_name":"Simulation","level":1,"score":0.2736999988555908},{"id":"https://openalex.org/C158154518","wikidata":"https://www.wikidata.org/wiki/Q7310970","display_name":"Relevance (law)","level":2,"score":0.27230000495910645},{"id":"https://openalex.org/C118021083","wikidata":"https://www.wikidata.org/wiki/Q610398","display_name":"System on a chip","level":2,"score":0.2660999894142151},{"id":"https://openalex.org/C94966114","wikidata":"https://www.wikidata.org/wiki/Q29256","display_name":"Black box","level":2,"score":0.26589998602867126},{"id":"https://openalex.org/C70437156","wikidata":"https://www.wikidata.org/wiki/Q7228652","display_name":"Pooling","level":2,"score":0.2655999958515167},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.2653000056743622},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C167085575","wikidata":"https://www.wikidata.org/wiki/Q6803654","display_name":"Mean squared prediction error","level":2,"score":0.2599000036716461},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.25760000944137573},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.2574999928474426},{"id":"https://openalex.org/C2779585090","wikidata":"https://www.wikidata.org/wiki/Q3457762","display_name":"Resilience (materials science)","level":2,"score":0.25130000710487366}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2605.23163","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.23163","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.2605.23163","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2605.23163","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":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"End-to-end":[0],"autonomous":[1],"driving":[2,78,160],"via":[3],"Vision-Language-Action":[4],"(VLA)":[5],"models":[6,39],"demands":[7,215],"a":[8,57,90,95,120,132,146],"precarious":[9],"balance":[10],"between":[11,209],"high-fidelity":[12],"trajectory":[13,129],"planning":[14],"and":[15,31,43,93,137,171,212],"efficient":[16],"inference.":[17],"Existing":[18],"paradigms":[19],"typically":[20],"fall":[21],"short:":[22],"autoregressive":[23],"(AR)":[24],"VLAs":[25,79,211],"are":[26],"memory-bandwidth-bound":[27],"on":[28,180],"edge":[29],"hardware":[30],"prone":[32],"to":[33,109,187],"exposure-bias":[34],"drift,":[35],"while":[36,67],"full-sequence":[37],"diffusion":[38],"preclude":[40],"KV-cache":[41],"reuse":[42],"suffer":[44],"from":[45,131],"\"logical":[46],"leakage\"":[47],"that":[48,60,77,99,153],"violates":[49],"the":[50,75,156,163,174,203,207,213],"fundamental":[51],"perceive-then-plan":[52],"causality.":[53],"We":[54,103],"present":[55],"Fast-dDrive,":[56],"block-diffusion":[58],"VLA":[59],"performs":[61],"bidirectional":[62],"refinement":[63],"within":[64],"semantic":[65],"units":[66],"enforcing":[68],"strict":[69],"causal":[70],"ordering":[71],"across":[72],"them.":[73],"Leveraging":[74],"observation":[76],"often":[80],"emit":[81],"structured":[82],"JSON-like":[83],"outputs,":[84],"Fast-dDrive":[85,154,167],"freezes":[86],"structural":[87],"tokens":[88],"into":[89],"section":[91],"scaffold":[92],"employs":[94],"section-aware":[96],"training":[97],"recipe":[98],"prioritizes":[100],"safety-critical":[101],"planning.":[102],"further":[104],"introduce":[105],"Scaffold":[106],"Speculative":[107],"Decoding":[108],"achieve":[110],"AR-equivalent":[111],"quality":[112],"at":[113,145],"significantly":[114],"higher":[115],"throughput.":[116],"Finally,":[117],"we":[118,140],"propose":[119],"low-overhead":[121],"test-time":[122],"scaling":[123],"scheme:":[124],"by":[125],"forking":[126],"$N$":[127],"stochastic":[128],"rollouts":[130],"single":[133],"shared-prefix":[134],"KV":[135],"cache":[136],"averaging":[138],"them,":[139],"effectively":[141],"suppress":[142],"prediction":[143],"variance":[144],"fractional":[147],"computational":[148],"cost.":[149],"Empirical":[150],"results":[151],"demonstrate":[152],"redefines":[155],"speed-accuracy":[157],"frontier":[158],"for":[159],"agents.":[161],"On":[162],"WOD-E2E":[164],"test":[165],"set,":[166],"achieves":[168],"SOTA":[169],"ADE@3s":[170],"ADE@5s,":[172],"alongside":[173],"highest":[175],"RFS":[176],"among":[177],"diffusion-based":[178],"VLAs;":[179],"nuScenes,":[181],"it":[182],"reduces":[183],"average":[184],"L2":[185],"error":[186],"$0.32$m":[188],"(a":[189],"$22\\%$":[190],"improvement).":[191],"When":[192],"integrated":[193],"with":[194],"SGLang,":[195],"our":[196],"framework":[197],"delivers":[198],"$12\\times$":[199],"throughput":[200],"speedup":[201],"over":[202],"AR":[204],"baseline,":[205],"narrowing":[206],"gap":[208],"high-capacity":[210],"efficiency":[214],"of":[216],"real-time":[217],"on-vehicle":[218],"deployment.":[219]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-05-26T00:00:00"}