{"id":"https://openalex.org/W7166541463","doi":"https://doi.org/10.48550/arxiv.2606.27660","title":"MVPruner: Dynamic Token Pruning for Accelerating Multi-view Vision-Language Models in Autonomous Driving","display_name":"MVPruner: Dynamic Token Pruning for Accelerating Multi-view Vision-Language Models in Autonomous Driving","publication_year":2026,"publication_date":"2026-06-26","ids":{"openalex":"https://openalex.org/W7166541463","doi":"https://doi.org/10.48550/arxiv.2606.27660"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.27660","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27660","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.2606.27660","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5139553954","display_name":"Nan Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Nan","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139623499","display_name":"Zhanwen Liu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Zhanwen","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139574275","display_name":"Linfeng Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Linfeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5068042573","display_name":"Shangyu Xie","orcid":"https://orcid.org/0000-0001-7484-6314"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xie, Shangyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139580868","display_name":"Yang Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, Yang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028993237","display_name":"Wenzhuo Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Wenzhuo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5053213848","display_name":"X Zhao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhao, Xiangmo","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/T11714","display_name":"Multimodal Machine Learning Applications","score":0.890999972820282,"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":0.890999972820282,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.03180000185966492,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.024800000712275505,"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/interpretability","display_name":"Interpretability","score":0.8169999718666077},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.7943000197410583},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.7455000281333923},{"id":"https://openalex.org/keywords/encode","display_name":"ENCODE","score":0.6241000294685364},{"id":"https://openalex.org/keywords/generalization","display_name":"Generalization","score":0.5702999830245972},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.5134000182151794},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.5077000260353088}],"concepts":[{"id":"https://openalex.org/C2781067378","wikidata":"https://www.wikidata.org/wiki/Q17027399","display_name":"Interpretability","level":2,"score":0.8169999718666077},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7996000051498413},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.7943000197410583},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7455000281333923},{"id":"https://openalex.org/C66746571","wikidata":"https://www.wikidata.org/wiki/Q1134833","display_name":"ENCODE","level":3,"score":0.6241000294685364},{"id":"https://openalex.org/C177148314","wikidata":"https://www.wikidata.org/wiki/Q170084","display_name":"Generalization","level":2,"score":0.5702999830245972},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5619999766349792},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.5134000182151794},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.5077000260353088},{"id":"https://openalex.org/C2780451532","wikidata":"https://www.wikidata.org/wiki/Q759676","display_name":"Task (project management)","level":2,"score":0.4562000036239624},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4553999900817871},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.35589998960494995},{"id":"https://openalex.org/C177769412","wikidata":"https://www.wikidata.org/wiki/Q278090","display_name":"Prior probability","level":3,"score":0.35249999165534973},{"id":"https://openalex.org/C97541855","wikidata":"https://www.wikidata.org/wiki/Q830687","display_name":"Reinforcement learning","level":2,"score":0.34439998865127563},{"id":"https://openalex.org/C94361409","wikidata":"https://www.wikidata.org/wiki/Q7882500","display_name":"Uncertainty reduction theory","level":2,"score":0.30730000138282776},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.2985000014305115},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.29750001430511475},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25699999928474426}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.27660","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27660","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.2606.27660","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.27660","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":{"Vision-Language":[0],"Models":[1],"(VLMs)":[2],"improve":[3],"generalization":[4],"and":[5,35,44,65,107,124],"interpretability":[6],"in":[7,22,62,156,160],"autonomous":[8],"driving":[9],"but":[10],"suffer":[11],"from":[12],"efficiency":[13],"issues":[14],"due":[15],"to":[16,131],"long":[17],"visual":[18],"token":[19,27,80],"sequences,":[20],"particularly":[21],"standard":[23],"multi-view":[24,55],"settings.":[25],"Existing":[26],"pruning":[28,32,81,85,97],"methods":[29],"employ":[30],"fixed":[31],"rate":[33],"allocation":[34],"static":[36],"importance":[37,42,48],"metrics,":[38],"ignoring":[39],"dynamic":[40,67,90],"inter-view":[41],"differences":[43],"the":[45,88,101,141],"evolving":[46],"information":[47,68,91,102],"during":[49],"inference.":[50],"Our":[51],"analysis":[52],"reveals":[53],"that":[54,83],"VLMs":[56],"inherently":[57],"encode":[58],"task-related":[59],"view":[60],"priors":[61],"deeper":[63],"layers":[64],"exhibit":[66],"requirements.":[69,92],"Motivated":[70],"by":[71,128],"these":[72],"findings,":[73],"we":[74],"propose":[75],"MVPruner,":[76],"a":[77],"two-stage":[78],"adaptive":[79],"method":[82],"aligns":[84],"behavior":[86],"with":[87,110,151],"model's":[89],"The":[93,119],"first":[94],"stage":[95,121],"allocates":[96,122],"budgets":[98,123],"based":[99],"on":[100,137,167],"diversity":[103],"of":[104,144],"each":[105],"view,":[106],"retains":[108],"tokens":[109,126],"consistent":[111],"contribution":[112],"across":[113],"stages,":[114],"ensuring":[115],"semantic":[116],"representational":[117],"capacity.":[118],"second":[120],"selects":[125],"guided":[127],"instruction":[129],"text":[130],"guarantee":[132],"task":[133],"alignment.":[134],"Experimental":[135],"results":[136],"four":[138],"benchmarks":[139],"demonstrate":[140],"superior":[142],"performance":[143],"our":[145],"method.":[146],"For":[147],"example,":[148],"DriveMM":[149],"equipped":[150],"MVPruner":[152],"achieves":[153],"87.3%":[154],"reduction":[155],"FLOPs,":[157],"4.97*":[158],"speedup":[159],"prefilling":[161],"phase":[162],"while":[163],"retaining":[164],"98.5%":[165],"accuracy":[166],"DriveLM":[168],"benchmark.":[169]},"counts_by_year":[],"updated_date":"2026-07-01T06:00:48.157686","created_date":"2026-06-30T00:00:00"}