{"id":"https://openalex.org/W7166827883","doi":"https://doi.org/10.48550/arxiv.2606.31382","title":"Revisiting Parameter Redundancy in Vision-Language-Action Models: Insights from VLM-to-VLA Adaptation","display_name":"Revisiting Parameter Redundancy in Vision-Language-Action Models: Insights from VLM-to-VLA Adaptation","publication_year":2026,"publication_date":"2026-06-30","ids":{"openalex":"https://openalex.org/W7166827883","doi":"https://doi.org/10.48550/arxiv.2606.31382"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2606.31382","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.31382","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":"Preprint"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2606.31382","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5047301622","display_name":"F Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Fengnian","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139831944","display_name":"Tao Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Huang, Tao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5053837667","display_name":"S Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Siyu","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139831559","display_name":"Zhong Jin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jin, Zhong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5139801665","display_name":"Chang Xu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Chang","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":0.7556999921798706,"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":0.7556999921798706,"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":0.037700001150369644,"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/T10883","display_name":"Ethics and Social Impacts of AI","score":0.018799999728798866,"subfield":{"id":"https://openalex.org/subfields/3311","display_name":"Safety Research"},"field":{"id":"https://openalex.org/fields/33","display_name":"Social Sciences"},"domain":{"id":"https://openalex.org/domains/2","display_name":"Social Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/redundancy","display_name":"Redundancy (engineering)","score":0.7484999895095825},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5773000121116638},{"id":"https://openalex.org/keywords/path","display_name":"Path (computing)","score":0.5009999871253967},{"id":"https://openalex.org/keywords/benchmark","display_name":"Benchmark (surveying)","score":0.4867999851703644},{"id":"https://openalex.org/keywords/divergence","display_name":"Divergence (linguistics)","score":0.4781999886035919},{"id":"https://openalex.org/keywords/adaptation","display_name":"Adaptation (eye)","score":0.47049999237060547},{"id":"https://openalex.org/keywords/sensitivity","display_name":"Sensitivity (control systems)","score":0.44589999318122864}],"concepts":[{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.7484999895095825},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5976999998092651},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5773000121116638},{"id":"https://openalex.org/C2777735758","wikidata":"https://www.wikidata.org/wiki/Q817765","display_name":"Path (computing)","level":2,"score":0.5009999871253967},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.4867999851703644},{"id":"https://openalex.org/C207390915","wikidata":"https://www.wikidata.org/wiki/Q1230525","display_name":"Divergence (linguistics)","level":2,"score":0.4781999886035919},{"id":"https://openalex.org/C139807058","wikidata":"https://www.wikidata.org/wiki/Q352374","display_name":"Adaptation (eye)","level":2,"score":0.47049999237060547},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.47049999237060547},{"id":"https://openalex.org/C21200559","wikidata":"https://www.wikidata.org/wiki/Q7451068","display_name":"Sensitivity (control systems)","level":2,"score":0.44589999318122864},{"id":"https://openalex.org/C101468663","wikidata":"https://www.wikidata.org/wiki/Q1620158","display_name":"Modular design","level":2,"score":0.4260999858379364},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3831999897956848},{"id":"https://openalex.org/C31170391","wikidata":"https://www.wikidata.org/wiki/Q188619","display_name":"Hierarchy","level":2,"score":0.34200000762939453},{"id":"https://openalex.org/C167928553","wikidata":"https://www.wikidata.org/wiki/Q1376021","display_name":"Estimation theory","level":2,"score":0.3244999945163727},{"id":"https://openalex.org/C63479239","wikidata":"https://www.wikidata.org/wiki/Q7353546","display_name":"Robustness (evolution)","level":3,"score":0.31139999628067017},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.28529998660087585},{"id":"https://openalex.org/C2778755073","wikidata":"https://www.wikidata.org/wiki/Q10858537","display_name":"Scale (ratio)","level":2,"score":0.28290000557899475},{"id":"https://openalex.org/C147764199","wikidata":"https://www.wikidata.org/wiki/Q6865248","display_name":"Minification","level":2,"score":0.2687999904155731},{"id":"https://openalex.org/C2983447341","wikidata":"https://www.wikidata.org/wiki/Q1413083","display_name":"Model parameter","level":2,"score":0.26089999079704285},{"id":"https://openalex.org/C126255220","wikidata":"https://www.wikidata.org/wiki/Q141495","display_name":"Mathematical optimization","level":1,"score":0.25529998540878296},{"id":"https://openalex.org/C33923547","wikidata":"https://www.wikidata.org/wiki/Q395","display_name":"Mathematics","level":0,"score":0.2529999911785126}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2606.31382","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.31382","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":"doi:10.48550/arxiv.2606.31382","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2606.31382","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":"Preprint"},"sustainable_development_goals":[{"score":0.41398385167121887,"display_name":"Peace, Justice and strong institutions","id":"https://metadata.un.org/sdg/16"},{"score":0.40149420499801636,"display_name":"Reduced inequalities","id":"https://metadata.un.org/sdg/10"}],"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":[2,34],"have":[3],"made":[4],"significant":[5],"strides":[6],"in":[7,207,225,239],"embodied":[8],"intelligence":[9],"by":[10,64,129,186],"integrating":[11],"the":[12,21,45,67,87,98,105,131,159,172,180,193,214,221],"powerful":[13],"representations":[14],"of":[15,25,90,100,108,134,182,192],"pre-trained":[16],"Vision-Language":[17],"Models":[18],"(VLMs).":[19],"However,":[20],"massive":[22],"parameter":[23,39,95,109,137,203,222],"scale":[24],"VLAs":[26],"imposes":[27],"a":[28,126,147,165,230],"heavy":[29],"computational":[30],"burden,":[31],"and":[32,154,184,228],"these":[33],"exhibit":[35],"extreme":[36],"sensitivity":[37],"to":[38,57,79,113],"pruning.":[40],"Current":[41],"paradigms":[42],"often":[43],"treat":[44],"resulting":[46],"performance":[47,77,141,195,209],"degradation":[48],"as":[49,125],"inevitable,":[50],"relying":[51],"on":[52,139,158,171],"fine-tuning":[53],"or":[54,82],"low-rank":[55],"corrections":[56],"recover":[58],"efficacy.":[59],"We":[60,93],"challenge":[61],"this":[62,84],"convention":[63],"questioning":[65],"whether":[66],"removed":[68],"parameters":[69,181],"are":[70],"truly":[71],"redundant":[72],"if":[73,83],"VLA":[74,140,226],"pruning":[75,89,124,168,204],"necessitates":[76],"recovery":[78],"be":[80],"effective,":[81],"paradigm":[85],"masks":[86],"indiscriminate":[88],"critical":[91],"parameters.":[92],"revisit":[94],"redundancy":[96],"through":[97],"lens":[99],"VLM-to-VLA":[101],"adaptation,":[102],"first":[103],"quantifying":[104],"spatial":[106],"distribution":[107],"divergence":[110,152],"during":[111],"adaptation":[112,227],"reveal":[114],"structured":[115],"patterns":[116],"across":[117],"different":[118,136],"modules.":[119],"Subsequently,":[120],"we":[121,145,163],"introduce":[122],"controlled":[123],"diagnostic":[127],"probe:":[128],"comparing":[130],"direct":[132],"impact":[133],"removing":[135],"subsets":[138],"without":[142,196],"any":[143,197],"fine-tuning,":[144],"establish":[146],"causal":[148],"link":[149],"between":[150],"adaptation-induced":[151],"signals":[153],"functional":[155],"contributions.":[156],"Based":[157],"discovered":[160],"modular":[161],"heterogeneities,":[162],"design":[164],"multi-module":[166],"joint":[167],"scheme.":[169],"Evaluations":[170],"LIBERO":[173],"benchmark":[174],"demonstrate":[175],"that":[176],"our":[177],"approach":[178],"reduces":[179],"OpenVLA":[183],"$\u03c0_{0.5}$":[185],"12\\%--30\\%":[187],"while":[188],"maintaining":[189],"approximately":[190],"90\\%":[191],"original":[194],"post-pruning":[198],"recovery.":[199],"In":[200],"contrast,":[201],"existing":[202],"criteria":[205],"result":[206],"total":[208],"collapse":[210],"when":[211],"evaluated":[212],"under":[213],"same":[215],"recovery-free":[216],"constraints.":[217],"Our":[218],"study":[219],"reveals":[220],"evolution":[223],"mechanism":[224],"provides":[229],"new":[231],"path":[232],"for":[233],"deploying":[234],"efficient,":[235],"robust":[236],"robotic":[237],"policies":[238],"resource-constrained":[240],"environments.":[241]},"counts_by_year":[],"updated_date":"2026-07-02T06:18:51.028212","created_date":"2026-07-02T00:00:00"}