{"id":"https://openalex.org/W7138192222","doi":"https://doi.org/10.1609/aaai.v40i4.37305","title":"StreamKV: Streaming Video Question-Answering with Segment-based KV Cache Retrieval and Compression","display_name":"StreamKV: Streaming Video Question-Answering with Segment-based KV Cache Retrieval and Compression","publication_year":2026,"publication_date":"2026-03-14","ids":{"openalex":"https://openalex.org/W7138192222","doi":"https://doi.org/10.1609/aaai.v40i4.37305"},"language":null,"primary_location":{"id":"doi:10.1609/aaai.v40i4.37305","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i4.37305","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"diamond","oa_url":"https://doi.org/10.1609/aaai.v40i4.37305","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5129715537","display_name":"Yilong Chen","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yilong Chen","raw_affiliation_strings":["State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University\nSchool of Software and Microelectronics, Peking University\nTaoBao & Tmall Group of Alibaba"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University\nSchool of Software and Microelectronics, Peking University\nTaoBao & Tmall Group of Alibaba","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129648730","display_name":"Xiang Bai","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiang Bai","raw_affiliation_strings":["State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University\nSchool of Software and Microelectronics, Peking University\nTaoBao & Tmall Group of Alibaba"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University\nSchool of Software and Microelectronics, Peking University\nTaoBao & Tmall Group of Alibaba","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129699982","display_name":"Zhibin Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zhibin Wang","raw_affiliation_strings":["TaoBao & Tmall Group of Alibaba"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"TaoBao & Tmall Group of Alibaba","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Chengyu Bai","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chengyu Bai","raw_affiliation_strings":["State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University\nSchool of Software and Microelectronics, Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University\nSchool of Software and Microelectronics, Peking University","institution_ids":["https://openalex.org/I20231570"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5129692992","display_name":"Yuhan Dai","orcid":null},"institutions":[{"id":"https://openalex.org/I4210095624","display_name":"Alibaba Group (United States)","ror":"https://ror.org/00rn0m335","country_code":"US","type":"company","lineage":["https://openalex.org/I4210095624","https://openalex.org/I45928872"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuhan Dai","raw_affiliation_strings":["TaoBao & Tmall Group of Alibaba"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"TaoBao & Tmall Group of Alibaba","institution_ids":["https://openalex.org/I4210095624"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5129709628","display_name":"Ming Lu","orcid":null},"institutions":[{"id":"https://openalex.org/I20231570","display_name":"Peking University","ror":"https://ror.org/02v51f717","country_code":"CN","type":"education","lineage":["https://openalex.org/I20231570"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ming Lu","raw_affiliation_strings":["State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"State Key Laboratory of Multimedia Information Processing, School of Computer Science, Peking University","institution_ids":["https://openalex.org/I20231570"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":43.3632,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.99604813,"is_in_top_1_percent":true,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":"40","issue":"4","first_page":"3120","last_page":"3128"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9779000282287598,"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":0.9779000282287598,"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/T11439","display_name":"Video Analysis and Summarization","score":0.00430000014603138,"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/T10028","display_name":"Topic Modeling","score":0.002300000051036477,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.8259000182151794},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.5342000126838684},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.43709999322891235},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.39649999141693115},{"id":"https://openalex.org/keywords/cpu-cache","display_name":"CPU cache","score":0.3637000024318695},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.3450999855995178},{"id":"https://openalex.org/keywords/video-retrieval","display_name":"Video retrieval","score":0.3433000147342682}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8862000107765198},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.8259000182151794},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.5342000126838684},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.43709999322891235},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.39649999141693115},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.37940001487731934},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.3776000142097473},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.3637000024318695},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.3450999855995178},{"id":"https://openalex.org/C2983174267","wikidata":"https://www.wikidata.org/wiki/Q3775098","display_name":"Video retrieval","level":2,"score":0.3433000147342682},{"id":"https://openalex.org/C25797200","wikidata":"https://www.wikidata.org/wiki/Q828137","display_name":"Compression ratio","level":3,"score":0.3375999927520752},{"id":"https://openalex.org/C38556500","wikidata":"https://www.wikidata.org/wiki/Q13404475","display_name":"Cache algorithms","level":4,"score":0.30820000171661377},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.2912999987602234},{"id":"https://openalex.org/C126831891","wikidata":"https://www.wikidata.org/wiki/Q221673","display_name":"Host (biology)","level":2,"score":0.27649998664855957},{"id":"https://openalex.org/C1667742","wikidata":"https://www.wikidata.org/wiki/Q10927554","display_name":"Image retrieval","level":3,"score":0.26969999074935913},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.26759999990463257},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.26080000400543213},{"id":"https://openalex.org/C2779304628","wikidata":"https://www.wikidata.org/wiki/Q3503480","display_name":"Face (sociological concept)","level":2,"score":0.2572000026702881}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1609/aaai.v40i4.37305","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i4.37305","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":{"id":"doi:10.1609/aaai.v40i4.37305","is_oa":true,"landing_page_url":"https://doi.org/10.1609/aaai.v40i4.37305","pdf_url":null,"source":{"id":"https://openalex.org/S4210191458","display_name":"Proceedings of the AAAI Conference on Artificial Intelligence","issn_l":"2159-5399","issn":["2159-5399","2374-3468"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/P4310320058","host_organization_name":"Association for the Advancement of Artificial Intelligence","host_organization_lineage":["https://openalex.org/P4310320058"],"host_organization_lineage_names":["Association for the Advancement of Artificial Intelligence"],"type":"conference"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the AAAI Conference on Artificial Intelligence","raw_type":"journal-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Video":[0],"Large":[1],"Language":[2],"Models":[3],"(Video-LLMs)":[4],"have":[5,30],"demonstrated":[6],"significant":[7],"potential":[8],"in":[9,170],"the":[10,44,53,137,146,177],"areas":[11],"of":[12,48,57,179],"video":[13,96,181],"captioning,":[14],"search,":[15],"and":[16,46,55,83,162,206],"summarization.":[17],"However,":[18,52],"current":[19],"Video-LLMs":[20,77],"still":[21,61],"face":[22],"challenges":[23],"with":[24,78],"long":[25,49],"real-world":[26,50],"videos.":[27,51],"Recent":[28],"methods":[29,88],"introduced":[31],"a":[32,71,112,131,165,171],"retrieval":[33,56,82,161],"mechanism":[34],"that":[35,74,89,190],"retrieves":[36],"query-relevant":[37],"KV":[38,58,80,107,126,149,159],"caches":[39,59,150],"for":[40,115,123,153],"question":[41,182],"answering,":[42],"enhancing":[43],"efficiency":[45,205],"accuracy":[47,199],"compression":[54,163],"are":[60,151],"not":[62],"fully":[63],"explored.":[64],"In":[65],"this":[66],"paper,":[67],"we":[68],"propose":[69],"StreamKV,":[70],"training-free":[72],"framework":[73],"seamlessly":[75],"equips":[76],"advanced":[79],"cache":[81,108,127,160],"compression.":[84],"Compared":[85],"to":[86,118,135],"previous":[87],"used":[90],"uniform":[91],"partitioning,":[92],"StreamKV":[93,110,129,157,191],"dynamically":[94],"partitions":[95],"streams":[97],"into":[98],"semantic":[99,104,139],"segments,":[100],"which":[101],"better":[102],"preserves":[103],"information.":[105],"For":[106,125],"retrieval,":[109],"calculates":[111],"summary":[113],"vector":[114],"each":[116,142],"segment":[117],"retain":[119],"segment-level":[120],"information":[121],"essential":[122],"retrieval.":[124],"compression,":[128],"introduces":[130],"guidance":[132],"prompt":[133],"designed":[134],"capture":[136],"key":[138],"elements":[140],"within":[141,164],"segment,":[143],"ensuring":[144],"only":[145],"most":[147],"informative":[148],"retained":[152],"answering":[154],"questions.":[155],"Moreover,":[156],"unifies":[158],"single":[166],"module,":[167],"performing":[168],"both":[169,203],"layer-adaptive":[172],"manner,":[173],"thereby":[174],"further":[175],"improving":[176,202],"effectiveness":[178],"streaming":[180],"answering.":[183],"Extensive":[184],"experiments":[185],"on":[186],"StreamingVQA":[187],"benchmarks":[188],"demonstrate":[189],"significantly":[192],"outperforms":[193],"existing":[194],"Online":[195],"Video-LLMs,":[196],"achieving":[197],"superior":[198],"while":[200],"substantially":[201],"memory":[204],"computational":[207],"latency.":[208]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-07-29T14:22:42.915294","created_date":"2026-03-18T00:00:00"}