{"id":"https://openalex.org/W4417102582","doi":"https://doi.org/10.1109/iccv51701.2025.02089","title":"Free-Moref: Instantly Multiplexing Context Perception Capabilities of Video-Mllms Within Single Inference","display_name":"Free-Moref: Instantly Multiplexing Context Perception Capabilities of Video-Mllms Within Single Inference","publication_year":2025,"publication_date":"2025-10-19","ids":{"openalex":"https://openalex.org/W4417102582","doi":"https://doi.org/10.1109/iccv51701.2025.02089"},"language":"en","primary_location":{"id":"doi:10.1109/iccv51701.2025.02089","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.02089","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref","datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2508.02134","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5073712729","display_name":"Kuo Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Kuo Wang","raw_affiliation_strings":["Sun Yat-sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sun Yat-sen University","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5038810646","display_name":"Quanlong Zheng","orcid":"https://orcid.org/0000-0001-5059-0078"},"institutions":[{"id":"https://openalex.org/I4210129149","display_name":"Postmaster Psychologie Opleidingen","ror":"https://ror.org/03naar428","country_code":"NL","type":"education","lineage":["https://openalex.org/I4210129149"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Quanlong Zheng","raw_affiliation_strings":["OPPO AI Center, OPPO Inc.,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"OPPO AI Center, OPPO Inc.,China","institution_ids":["https://openalex.org/I4210129149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5014208479","display_name":"Junlin Xie","orcid":null},"institutions":[{"id":"https://openalex.org/I4210116924","display_name":"Chinese University of Hong Kong, Shenzhen","ror":"https://ror.org/02d5ks197","country_code":"CN","type":"education","lineage":["https://openalex.org/I177725633","https://openalex.org/I180726961","https://openalex.org/I4210116924"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junlin Xie","raw_affiliation_strings":["The Chinese University of Hong Kong,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"The Chinese University of Hong Kong,Shenzhen,China","institution_ids":["https://openalex.org/I4210116924"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5118216947","display_name":"Yanhao Zhang","orcid":"https://orcid.org/0000-0002-6239-8953"},"institutions":[{"id":"https://openalex.org/I4210129149","display_name":"Postmaster Psychologie Opleidingen","ror":"https://ror.org/03naar428","country_code":"NL","type":"education","lineage":["https://openalex.org/I4210129149"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Yanhao Zhang","raw_affiliation_strings":["OPPO AI Center, OPPO Inc.,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"OPPO AI Center, OPPO Inc.,China","institution_ids":["https://openalex.org/I4210129149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5001147139","display_name":"Jinguo Luo","orcid":"https://orcid.org/0009-0007-4069-1984"},"institutions":[{"id":"https://openalex.org/I204983213","display_name":"Harbin Institute of Technology","ror":"https://ror.org/01yqg2h08","country_code":"CN","type":"education","lineage":["https://openalex.org/I204983213"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Jinguo Luo","raw_affiliation_strings":["Harbin Institute of Technology,Shenzhen,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Harbin Institute of Technology,Shenzhen,China","institution_ids":["https://openalex.org/I204983213"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5007720098","display_name":"H. Lu","orcid":"https://orcid.org/0000-0001-6332-2785"},"institutions":[{"id":"https://openalex.org/I4210129149","display_name":"Postmaster Psychologie Opleidingen","ror":"https://ror.org/03naar428","country_code":"NL","type":"education","lineage":["https://openalex.org/I4210129149"]}],"countries":["NL"],"is_corresponding":false,"raw_author_name":"Haonan Lu","raw_affiliation_strings":["OPPO AI Center, OPPO Inc.,China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"OPPO AI Center, OPPO Inc.,China","institution_ids":["https://openalex.org/I4210129149"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5087020613","display_name":"Lin Liang","orcid":"https://orcid.org/0000-0002-4879-4642"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Liang Lin","raw_affiliation_strings":["Sun Yat-sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sun Yat-sen University","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5090896025","display_name":"Fan Zhou","orcid":"https://orcid.org/0000-0002-0400-9366"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fan Zhou","raw_affiliation_strings":["Sun Yat-sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sun Yat-sen University","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5042965510","display_name":"Guanbin Li","orcid":"https://orcid.org/0000-0002-4805-0926"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Guanbin Li","raw_affiliation_strings":["Sun Yat-sen University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Sun Yat-sen University","institution_ids":["https://openalex.org/I157773358"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"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":"22499","last_page":"22508"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.8235999941825867,"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":0.8235999941825867,"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":0.03720000013709068,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.03620000183582306,"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":0.7581999897956848},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.7207000255584717},{"id":"https://openalex.org/keywords/context","display_name":"Context (archaeology)","score":0.6029000282287598},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.4772999882698059},{"id":"https://openalex.org/keywords/multiplexing","display_name":"Multiplexing","score":0.47130000591278076},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.47099998593330383},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.4528999924659729},{"id":"https://openalex.org/keywords/frame","display_name":"Frame (networking)","score":0.43309998512268066},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.39959999918937683}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8338000178337097},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.7581999897956848},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7207000255584717},{"id":"https://openalex.org/C2779343474","wikidata":"https://www.wikidata.org/wiki/Q3109175","display_name":"Context (archaeology)","level":2,"score":0.6029000282287598},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5360000133514404},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.4772999882698059},{"id":"https://openalex.org/C19275194","wikidata":"https://www.wikidata.org/wiki/Q222903","display_name":"Multiplexing","level":2,"score":0.47130000591278076},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.47099998593330383},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.4528999924659729},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.43309998512268066},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4083000123500824},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.39959999918937683},{"id":"https://openalex.org/C183322885","wikidata":"https://www.wikidata.org/wiki/Q17007702","display_name":"Context model","level":3,"score":0.3903999924659729},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.38370001316070557},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.36980000138282776},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.36070001125335693},{"id":"https://openalex.org/C57273362","wikidata":"https://www.wikidata.org/wiki/Q576722","display_name":"Decoding methods","level":2,"score":0.33309999108314514},{"id":"https://openalex.org/C117797892","wikidata":"https://www.wikidata.org/wiki/Q286363","display_name":"Shadow (psychology)","level":2,"score":0.3215999901294708},{"id":"https://openalex.org/C43521106","wikidata":"https://www.wikidata.org/wiki/Q2165493","display_name":"Pipeline (software)","level":2,"score":0.3122999966144562},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3057999908924103},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.3005000054836273},{"id":"https://openalex.org/C172849965","wikidata":"https://www.wikidata.org/wiki/Q3148875","display_name":"Reference frame","level":3,"score":0.2865000069141388},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.27000001072883606},{"id":"https://openalex.org/C184898388","wikidata":"https://www.wikidata.org/wiki/Q1435712","display_name":"Pairwise comparison","level":2,"score":0.26499998569488525},{"id":"https://openalex.org/C5339829","wikidata":"https://www.wikidata.org/wiki/Q1425977","display_name":"Machine vision","level":2,"score":0.263700008392334},{"id":"https://openalex.org/C3261483","wikidata":"https://www.wikidata.org/wiki/Q119565","display_name":"Frame rate","level":2,"score":0.26249998807907104},{"id":"https://openalex.org/C185798385","wikidata":"https://www.wikidata.org/wiki/Q1161707","display_name":"Benchmark (surveying)","level":2,"score":0.26100000739097595},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.258899986743927},{"id":"https://openalex.org/C28490314","wikidata":"https://www.wikidata.org/wiki/Q189436","display_name":"Speech recognition","level":1,"score":0.2574999928474426},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.2531999945640564},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.2508000135421753}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/iccv51701.2025.02089","is_oa":false,"landing_page_url":"https://doi.org/10.1109/iccv51701.2025.02089","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE/CVF International Conference on Computer Vision (ICCV)","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2508.02134","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.02134","pdf_url":"https://arxiv.org/pdf/2508.02134","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"doi:10.48550/arxiv.2508.02134","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2508.02134","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":"pmh:oai:arXiv.org:2508.02134","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.02134","pdf_url":"https://arxiv.org/pdf/2508.02134","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":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1437690383","display_name":null,"funder_award_id":"2024YFB3908503","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"},{"id":"https://openalex.org/G4751043524","display_name":null,"funder_award_id":"62322608","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Video":[0],"Multimodal":[1],"Large":[2],"Language":[3],"Models~(Video-MLLM)":[4],"have":[5],"achieved":[6],"remarkable":[7],"advancements":[8],"in":[9,20,112,123,153,173],"video":[10,32],"understanding":[11,60],"tasks.":[12],"However,":[13],"constrained":[14],"by":[15],"the":[16,21,79,92,109,120,146,159],"context":[17,80,176],"length":[18],"limitation":[19],"underlying":[22],"LLMs,":[23,124],"existing":[24],"Video-MLLMs":[25,84],"typically":[26],"exhibit":[27],"suboptimal":[28],"performance":[29,167,215],"on":[30,184,203],"long":[31,160],"scenarios.":[33],"To":[34],"understand":[35],"extended":[36],"input":[37,199],"frames,":[38],"common":[39],"solutions":[40],"span":[41],"token":[42,162],"compression":[43,202],"and":[44,70,157,181],"streaming":[45],"inference":[46,53,87],"techniques,":[47],"which":[48,76,105,144],"sacrifice":[49],"feature":[50],"granularity":[51],"or":[52],"efficiency.":[54],"Differently,":[55],"to":[56,114,131,196],"efficiently":[57],"achieve":[58],"comprehensive":[59],"of":[61,83,194],"longer":[62,198],"frame":[63],"inputs,":[64],"we":[65,102],"draw":[66],"ideas":[67],"from":[68,108,141],"MoE":[69],"propose":[71],"a":[72,125,133,204],"training-free":[73],"approach":[74],"\\textbf{Free-MoRef},":[75],"instantly":[77],"multiplexes":[78],"perception":[81,193],"capabilities":[82],"within":[85],"one":[86],"pass.":[88],"Specifically,":[89],"Free-MoRef":[90,164,190],"reconstructs":[91],"vision":[93,148,161],"tokens":[94,140],"into":[95],"several":[96],"short":[97],"sequences":[98],"as":[99],"multi-references.":[100],"Subsequently,":[101],"introduce":[103],"MoRef-attention,":[104],"gathers":[106],"clues":[107],"multi-reference":[110],"chunks":[111],"parallel":[113,142],"summarize":[115],"unified":[116],"query":[117],"activations.":[118],"After":[119],"shadow":[121],"layers":[122],"reference":[126],"fusion":[127],"step":[128],"is":[129],"derived":[130],"compose":[132],"final":[134],"mixed":[135],"reasoning":[136,174],"sequence":[137],"with":[138],"key":[139],"chunks,":[143],"compensates":[145],"cross-reference":[147],"interactions":[149],"that":[150,189],"are":[151,223],"neglected":[152],"MoRef-attention.":[154],"By":[155],"splitting":[156],"fusing":[158],"sequences,":[163],"achieves":[165,191],"improved":[166],"under":[168],"much":[169],"lower":[170],"computing":[171],"costs":[172],"multiplexed":[175],"length,":[177],"demonstrating":[178],"strong":[179],"efficiency":[180],"effectiveness.":[182],"Experiments":[183],"VideoMME,":[185],"MLVU,":[186],"LongVideoBench":[187],"show":[188],"full":[192],"2$\\times$":[195],"8$\\times$":[197],"frames":[200],"without":[201],"single":[205],"A100":[206],"GPU":[207],"while":[208],"keeping":[209],"instant":[210],"responses,":[211],"thereby":[212],"bringing":[213],"significant":[214],"gains,":[216],"even":[217],"surpassing":[218],"dedicatedly":[219],"trained":[220],"long-video-MLLMs.":[221],"Codes":[222],"available":[224],"at":[225],"https://github.com/wkfdb/Free-MoRef":[226]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}