{"id":"https://openalex.org/W7140308850","doi":"https://doi.org/10.48550/arxiv.2603.22911","title":"ForestPrune: High-ratio Visual Token Compression for Video Multimodal Large Language Models via Spatial-Temporal Forest Modeling","display_name":"ForestPrune: High-ratio Visual Token Compression for Video Multimodal Large Language Models via Spatial-Temporal Forest Modeling","publication_year":2026,"publication_date":"2026-03-24","ids":{"openalex":"https://openalex.org/W7140308850","doi":"https://doi.org/10.48550/arxiv.2603.22911"},"language":null,"primary_location":{"id":"doi:10.48550/arxiv.2603.22911","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.22911","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":"article"},"type":"preprint","indexed_in":["datacite"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://doi.org/10.48550/arxiv.2603.22911","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5130553450","display_name":"Shaobo Ju","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ju, Shaobo","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5125070953","display_name":"Baiyang Song","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Song, Baiyang","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130546135","display_name":"Tao Chen","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chen, Tao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130541835","display_name":"Jiapeng Zhang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhang, Jiapeng","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130592128","display_name":"Qiong Wu","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wu, Qiong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122033995","display_name":"Chao Chang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chang, Chao","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5122070586","display_name":"HuaiXi Wang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Wang, HuaiXi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"middle","author":{"id":"https://openalex.org/A5130576018","display_name":"Yiyi Zhou","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhou, Yiyi","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]},{"author_position":"last","author":{"id":"https://openalex.org/A5130597579","display_name":"Rongrong Ji","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Ji, Rongrong","raw_affiliation_strings":[],"raw_orcid":null,"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":9,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.23549999296665192,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.23549999296665192,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.156700000166893,"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/T12357","display_name":"Digital Media Forensic Detection","score":0.07320000231266022,"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":0.8395000100135803},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.6917999982833862},{"id":"https://openalex.org/keywords/tree","display_name":"Tree (set theory)","score":0.46000000834465027},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.45410001277923584},{"id":"https://openalex.org/keywords/construct","display_name":"Construct (python library)","score":0.4246000051498413},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.41519999504089355}],"concepts":[{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.8395000100135803},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8361999988555908},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.6917999982833862},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5138999819755554},{"id":"https://openalex.org/C113174947","wikidata":"https://www.wikidata.org/wiki/Q2859736","display_name":"Tree (set theory)","level":2,"score":0.46000000834465027},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.45410001277923584},{"id":"https://openalex.org/C2780801425","wikidata":"https://www.wikidata.org/wiki/Q5164392","display_name":"Construct (python library)","level":2,"score":0.4246000051498413},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.41519999504089355},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.37940001487731934},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.36000001430511475},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.34709998965263367},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.31529998779296875},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3075000047683716},{"id":"https://openalex.org/C62611344","wikidata":"https://www.wikidata.org/wiki/Q1062658","display_name":"Node (physics)","level":2,"score":0.2985000014305115},{"id":"https://openalex.org/C115067241","wikidata":"https://www.wikidata.org/wiki/Q1639854","display_name":"Token passing","level":3,"score":0.28369998931884766},{"id":"https://openalex.org/C163797641","wikidata":"https://www.wikidata.org/wiki/Q2067937","display_name":"Tree structure","level":3,"score":0.25429999828338623},{"id":"https://openalex.org/C161765866","wikidata":"https://www.wikidata.org/wiki/Q184748","display_name":"Codec","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2603.22911","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.22911","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":"article"}],"best_oa_location":{"id":"doi:10.48550/arxiv.2603.22911","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2603.22911","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":"article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/15","score":0.6550047397613525,"display_name":"Life in Land"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Due":[0],"to":[1,43,130],"the":[2,28,44,88,103,155,181],"great":[3,156],"saving":[4],"of":[5,35,47,98,105,146],"computation":[6],"and":[7,19,49,53,57,70,91,108,114,137,139,178,191],"memory":[8],"overhead,":[9],"token":[10,37,59,81,106,183],"compression":[11,184],"has":[12],"become":[13],"a":[14,55,119,144],"research":[15],"hot-spot":[16],"for":[17,27,62,158,170],"MLLMs":[18],"achieved":[20],"remarkable":[21],"progress":[22],"in":[23],"image-language":[24],"tasks.":[25],"However,":[26],"video,":[29],"existing":[30],"methods":[31],"still":[32],"fall":[33],"short":[34],"high-ratio":[36,71],"compression.":[38],"We":[39],"attribute":[40],"this":[41],"shortcoming":[42],"insufficient":[45],"modeling":[46],"temporal":[48,92],"continual":[50],"video":[51,63,84,133,147,159],"content,":[52],"propose":[54],"novel":[56],"training-free":[58],"pruning":[60,72,122,193],"method":[61],"MLLMs,":[64,134,160],"termed":[65],"ForestPrune,":[66,126],"which":[67],"achieves":[68],"effective":[69],"via":[73],"Spatial-temporal":[74],"Forest":[75],"Modeling.":[76],"In":[77],"practice,":[78],"ForestPrune":[79,101],"construct":[80],"forests":[82],"across":[83],"frames":[85],"based":[86,110],"on":[87,111,143,189,197],"semantic,":[89],"spatial":[90],"constraints,":[93],"making":[94],"an":[95],"overall":[96],"comprehension":[97],"videos.":[99],"Afterwards,":[100],"evaluates":[102],"importance":[104],"trees":[107],"nodes":[109],"tree":[112],"depth":[113],"node":[115],"roles,":[116],"thereby":[117],"obtaining":[118],"globally":[120],"optimal":[121],"decision.":[123],"To":[124],"validate":[125],"we":[127],"apply":[128],"it":[129],"two":[131],"representative":[132],"namely":[135],"LLaVA-Video":[136],"LLaVA-OneVision,":[138,171],"conduct":[140],"extensive":[141],"experiments":[142],"bunch":[145],"benchmarks.":[148],"The":[149],"experimental":[150],"results":[151],"not":[152],"only":[153],"show":[154,174],"effectiveness":[157],"e.g.,":[161,186],"retaining":[162],"95.8%":[163],"average":[164],"accuracy":[165,188],"while":[166],"reducing":[167],"90%":[168],"tokens":[169],"but":[172],"also":[173],"its":[175],"superior":[176],"performance":[177],"efficiency":[179],"than":[180,195],"compared":[182],"methods,":[185],"+10.1%":[187],"MLVU":[190],"-81.4%":[192],"time":[194],"FrameFusion":[196],"LLaVA-Video.":[198]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-03-26T00:00:00"}