{"id":"https://openalex.org/W6891792589","doi":"https://doi.org/10.48550/arxiv.2504.15270","title":"An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes","display_name":"An LMM for Efficient Video Understanding via Reinforced Compression of Video Cubes","publication_year":2025,"publication_date":"2025-04-21","ids":{"openalex":"https://openalex.org/W6891792589","doi":"https://doi.org/10.48550/arxiv.2504.15270"},"language":"en","primary_location":{"id":"doi:10.48550/arxiv.2504.15270","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.15270","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.2504.15270","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Qi, Ji","orcid":null},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Qi, Ji","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Yao, Yuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yao, Yuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Bai, Yushi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Bai, Yushi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Xu, Bin","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xu, Bin","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Li, Juanzi","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Li, Juanzi","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"middle","author":{"id":null,"display_name":"Liu, Zhiyuan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liu, Zhiyuan","raw_affiliation_strings":[],"affiliations":[]},{"author_position":"last","author":{"id":null,"display_name":"Chua, Tat-Seng","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Chua, Tat-Seng","raw_affiliation_strings":[],"affiliations":[]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":7,"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":true,"primary_topic":{"id":"https://openalex.org/T10901","display_name":"Advanced Data Compression Techniques","score":0.16820000112056732,"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/T10901","display_name":"Advanced Data Compression Techniques","score":0.16820000112056732,"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/T10860","display_name":"Speech and Audio Processing","score":0.14730000495910645,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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":0.14470000565052032,"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/redundancy","display_name":"Redundancy (engineering)","score":0.611299991607666},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.5770999789237976},{"id":"https://openalex.org/keywords/resampling","display_name":"Resampling","score":0.48429998755455017},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.4284000098705292},{"id":"https://openalex.org/keywords/frame","display_name":"Frame (networking)","score":0.4268999993801117},{"id":"https://openalex.org/keywords/cube","display_name":"Cube (algebra)","score":0.39160001277923584},{"id":"https://openalex.org/keywords/sequence","display_name":"Sequence (biology)","score":0.3822000026702881},{"id":"https://openalex.org/keywords/video-compression-picture-types","display_name":"Video compression picture types","score":0.38109999895095825}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7885000109672546},{"id":"https://openalex.org/C152124472","wikidata":"https://www.wikidata.org/wiki/Q1204361","display_name":"Redundancy (engineering)","level":2,"score":0.611299991607666},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.5770999789237976},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5080999732017517},{"id":"https://openalex.org/C150921843","wikidata":"https://www.wikidata.org/wiki/Q1170431","display_name":"Resampling","level":2,"score":0.48429998755455017},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.4514999985694885},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.4284000098705292},{"id":"https://openalex.org/C126042441","wikidata":"https://www.wikidata.org/wiki/Q1324888","display_name":"Frame (networking)","level":2,"score":0.4268999993801117},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.40639999508857727},{"id":"https://openalex.org/C53051483","wikidata":"https://www.wikidata.org/wiki/Q861555","display_name":"Cube (algebra)","level":2,"score":0.39160001277923584},{"id":"https://openalex.org/C2778112365","wikidata":"https://www.wikidata.org/wiki/Q3511065","display_name":"Sequence (biology)","level":2,"score":0.3822000026702881},{"id":"https://openalex.org/C106030495","wikidata":"https://www.wikidata.org/wiki/Q1797012","display_name":"Video compression picture types","level":4,"score":0.38109999895095825},{"id":"https://openalex.org/C99844830","wikidata":"https://www.wikidata.org/wiki/Q102441924","display_name":"Scaling","level":2,"score":0.3772999942302704},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.35670000314712524},{"id":"https://openalex.org/C774472","wikidata":"https://www.wikidata.org/wiki/Q6760393","display_name":"Margin (machine learning)","level":2,"score":0.3538999855518341},{"id":"https://openalex.org/C2778869765","wikidata":"https://www.wikidata.org/wiki/Q6028363","display_name":"Inefficiency","level":2,"score":0.3474000096321106},{"id":"https://openalex.org/C137800194","wikidata":"https://www.wikidata.org/wiki/Q11713455","display_name":"Interpolation (computer graphics)","level":3,"score":0.29679998755455017},{"id":"https://openalex.org/C65483669","wikidata":"https://www.wikidata.org/wiki/Q3536669","display_name":"Video processing","level":2,"score":0.29409998655319214},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.2851000130176544},{"id":"https://openalex.org/C3261483","wikidata":"https://www.wikidata.org/wiki/Q119565","display_name":"Frame rate","level":2,"score":0.2786000072956085},{"id":"https://openalex.org/C196496179","wikidata":"https://www.wikidata.org/wiki/Q860741","display_name":"Group of pictures","level":3,"score":0.27489998936653137},{"id":"https://openalex.org/C2780586882","wikidata":"https://www.wikidata.org/wiki/Q7520643","display_name":"Simple (philosophy)","level":2,"score":0.27000001072883606},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.26260000467300415},{"id":"https://openalex.org/C198531522","wikidata":"https://www.wikidata.org/wiki/Q485146","display_name":"Sample (material)","level":2,"score":0.25949999690055847},{"id":"https://openalex.org/C110384440","wikidata":"https://www.wikidata.org/wiki/Q1143270","display_name":"Upsampling","level":3,"score":0.2567000091075897}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.48550/arxiv.2504.15270","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.15270","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.2504.15270","is_oa":true,"landing_page_url":"https://doi.org/10.48550/arxiv.2504.15270","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":[{"display_name":"Peace, Justice and strong institutions","score":0.4316307008266449,"id":"https://metadata.un.org/sdg/16"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Multimodal":[1],"Models":[2],"(LMMs)":[3],"uniformly":[4],"perceive":[5],"video":[6,32,53,62],"frames,":[7],"creating":[8],"computational":[9],"inefficiency":[10],"for":[11,47,116,196],"videos":[12,100],"with":[13,25,81],"inherently":[14],"varying":[15,37],"temporal":[16,67],"information":[17],"density.":[18],"This":[19,55],"paper":[20],"present":[21],"\\textbf{Quicksviewer},":[22],"an":[23],"LMM":[24],"new":[26],"perceiving":[27,108],"paradigm":[28],"that":[29,186],"partitions":[30],"a":[31,44,90,125,130,173],"of":[33,103,132,155,169,177],"nonuniform":[34],"density":[35],"into":[36],"cubes":[38],"using":[39,150],"Gumbel":[40],"Softmax,":[41],"followed":[42],"by":[43,129,160,190],"unified":[45],"resampling":[46],"each":[48,97],"cube":[49],"to":[50,106,153],"achieve":[51],"efficient":[52,79],"understanding.":[54],"simple":[56],"and":[57],"intuitive":[58],"approach":[59],"dynamically":[60],"compress":[61],"online":[63],"based":[64],"on":[65,101],"its":[66],"density,":[68],"significantly":[69],"reducing":[70],"spatiotemporal":[71],"redundancy":[72],"(overall":[73],"45$\\times$":[74],"compression":[75],"rate),":[76],"while":[77],"enabling":[78],"training":[80],"large":[82],"receptive":[83],"field.":[84],"We":[85],"train":[86],"the":[87,107,121,137,167,178,187,191],"model":[88,119,179],"from":[89],"language":[91],"backbone":[92],"through":[93],"three":[94],"progressive":[95],"stages,":[96],"incorporating":[98],"lengthy":[99],"average":[102],"420s/1fps":[104],"thanks":[105],"efficiency.":[109],"With":[110,162],"only":[111],"0.8M":[112],"total":[113],"video-text":[114],"samples":[115],"training,":[117],"our":[118],"outperforms":[120],"direct":[122],"baseline":[123],"employing":[124],"fixed":[126],"partitioning":[127],"strategy":[128],"maximum":[131],"8.72":[133],"in":[134,139,200],"accuracy,":[135],"demonstrating":[136],"effectiveness":[138],"performance.":[140],"On":[141],"Video-MME,":[142],"Quicksviewer":[143],"achieves":[144],"SOTA":[145],"under":[146],"modest":[147],"sequence":[148],"lengths":[149],"just":[151],"up":[152,166],"5\\%":[154],"tokens":[156],"per":[157],"frame":[158],"required":[159],"baselines.":[161],"this":[163],"paradigm,":[164],"scaling":[165],"number":[168],"input":[170],"frames":[171],"reveals":[172],"clear":[174],"power":[175],"law":[176],"capabilities.":[180],"It":[181],"is":[182],"also":[183],"empirically":[184],"verified":[185],"segments":[188],"generated":[189],"cubing":[192],"network":[193],"can":[194],"help":[195],"analyzing":[197],"continuous":[198],"events":[199],"videos.":[201]},"counts_by_year":[],"updated_date":"2025-11-06T06:51:31.235846","created_date":"2025-10-10T00:00:00"}