{"id":"https://openalex.org/W7150982893","doi":"https://doi.org/10.48550/arxiv.2604.02689","title":"Efficient3D: A Unified Framework for Adaptive and Debiased Token Reduction in 3D MLLMs","display_name":"Efficient3D: A Unified Framework for Adaptive and Debiased Token Reduction in 3D MLLMs","publication_year":2026,"publication_date":"2026-04-03","ids":{"openalex":"https://openalex.org/W7150982893","doi":"https://doi.org/10.48550/arxiv.2604.02689"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2604.02689","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.02689","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2604.02689","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5133044708","display_name":"Yuhui Lin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Lin, Yuhui","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133024677","display_name":"Siyue Yu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yu, Siyue","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133033412","display_name":"Yuxing Yang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yang, Yuxing","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5133058385","display_name":"Guangliang Cheng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Cheng, Guangliang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5133055201","display_name":"Jimin Xiao","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiao, Jimin","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":0.7063999772071838,"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":0.7063999772071838,"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":0.10379999876022339,"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":0.031700000166893005,"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/security-token","display_name":"Security token","score":0.7342000007629395},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5566999912261963},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.5286999940872192},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5257999897003174},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.4887000024318695},{"id":"https://openalex.org/keywords/estimator","display_name":"Estimator","score":0.4097999930381775},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.37229999899864197},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.3578000068664551}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7839999794960022},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7342000007629395},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5566999912261963},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.5286999940872192},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5257999897003174},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4902999997138977},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.4887000024318695},{"id":"https://openalex.org/C185429906","wikidata":"https://www.wikidata.org/wiki/Q1130160","display_name":"Estimator","level":2,"score":0.4097999930381775},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.37229999899864197},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.36500000953674316},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.3578000068664551},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3319999873638153},{"id":"https://openalex.org/C153083717","wikidata":"https://www.wikidata.org/wiki/Q6535263","display_name":"Leverage (statistics)","level":2,"score":0.33070001006126404},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.32739999890327454},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.3246999979019165},{"id":"https://openalex.org/C206345919","wikidata":"https://www.wikidata.org/wiki/Q20380951","display_name":"Resource (disambiguation)","level":2,"score":0.3093999922275543},{"id":"https://openalex.org/C70518039","wikidata":"https://www.wikidata.org/wiki/Q16000077","display_name":"Dimensionality reduction","level":2,"score":0.2863999903202057},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.2851000130176544},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.26089999079704285},{"id":"https://openalex.org/C17231256","wikidata":"https://www.wikidata.org/wiki/Q5156540","display_name":"Completeness (order theory)","level":2,"score":0.26080000400543213},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.2590000033378601},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.25209999084472656}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2604.02689","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.02689","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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.2604.02689","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2604.02689","pdf_url":null,"source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"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":{"Recent":[0],"advances":[1],"in":[2,189],"Multimodal":[3],"Large":[4],"Language":[5],"Models":[6],"(MLLMs)":[7],"have":[8],"expanded":[9],"reasoning":[10],"capabilities":[11],"into":[12],"3D":[13,24,62,147,190],"domains,":[14],"enabling":[15],"fine-grained":[16],"spatial":[17],"understanding.":[18],"However,":[19],"the":[20,27,82,175],"substantial":[21],"size":[22],"of":[23,30,84],"MLLMs":[25,63],"and":[26,122,149,157,183],"high":[28],"dimensionality":[29],"input":[31],"features":[32],"introduce":[33],"considerable":[34],"inference":[35,188],"overhead,":[36],"which":[37,80],"limits":[38],"practical":[39],"deployment":[40],"on":[41,116,144,174],"resource":[42],"constrained":[43],"platforms.":[44],"To":[45],"overcome":[46],"this":[47,49],"limitation,":[48],"paper":[50],"presents":[51],"Efficient3D,":[52],"a":[53,72,170,181],"unified":[54],"framework":[55,70],"for":[56,97,186],"visual":[57,98],"token":[58,132],"pruning":[59,113],"that":[60,134,160],"accelerates":[61],"while":[64],"maintaining":[65,123],"competitive":[66],"accuracy.":[67],"The":[68,192],"proposed":[69],"introduces":[71],"Debiased":[73],"Visual":[74],"Token":[75,104],"Importance":[76],"Estimator":[77],"(DVTIE)":[78],"module,":[79],"considers":[81],"influence":[83],"shallow":[85],"initial":[86],"layers":[87],"during":[88],"attention":[89,125],"aggregation,":[90],"thereby":[91],"producing":[92],"more":[93],"reliable":[94],"importance":[95],"predictions":[96],"tokens.":[99],"In":[100],"addition,":[101],"an":[102],"Adaptive":[103],"Rebalancing":[105],"(ATR)":[106],"strategy":[107],"is":[108,194],"developed":[109],"to":[110],"dynamically":[111],"adjust":[112],"strength":[114],"based":[115],"scene":[117],"complexity,":[118],"preserving":[119],"semantic":[120],"completeness":[121],"balanced":[124],"across":[126],"layers.":[127],"Together,":[128],"they":[129],"enable":[130],"context-aware":[131],"reduction":[133],"maintains":[135],"essential":[136],"semantics":[137],"with":[138,166,169],"lower":[139],"computation.":[140],"Comprehensive":[141],"experiments":[142],"conducted":[143],"five":[145],"representative":[146],"vision":[148],"language":[150],"benchmarks,":[151],"including":[152],"ScanRefer,":[153],"Multi3DRefer,":[154],"Scan2Cap,":[155],"ScanQA,":[156],"SQA3D,":[158],"demonstrate":[159],"Efficient3D":[161,179],"achieves":[162],"superior":[163],"performance":[164],"compared":[165],"unpruned":[167],"baselines,":[168],"+2.57%":[171],"CIDEr":[172],"improvement":[173],"Scan2Cap":[176],"dataset.":[177],"Therefore,":[178],"provides":[180],"scalable":[182],"effective":[184],"solution":[185],"efficient":[187],"MLLMs.":[191],"code":[193],"released":[195],"at:":[196],"https://github.com/sol924/Efficient3D":[197]},"counts_by_year":[],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-04-07T00:00:00"}