{"id":"https://openalex.org/W4408351911","doi":"https://doi.org/10.1109/icassp49660.2025.10890678","title":"Find Details in Long Videos: Tower-of-Thoughts and Self-Retrieval Augmented Generation for Video Understanding","display_name":"Find Details in Long Videos: Tower-of-Thoughts and Self-Retrieval Augmented Generation for Video Understanding","publication_year":2025,"publication_date":"2025-03-12","ids":{"openalex":"https://openalex.org/W4408351911","doi":"https://doi.org/10.1109/icassp49660.2025.10890678"},"language":"en","primary_location":{"id":"doi:10.1109/icassp49660.2025.10890678","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890678","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":false,"oa_status":"closed","oa_url":null,"any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5068996540","display_name":"Tong Yue","orcid":"https://orcid.org/0000-0002-3330-3468"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Tong Yue","raw_affiliation_strings":["Tsinghua University,Department of Electronic Engineering,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Electronic Engineering,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108993608","display_name":"Mingrui Xiao","orcid":null},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Mingrui Xiao","raw_affiliation_strings":["Tsinghua University,Department of Electronic Engineering,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Electronic Engineering,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5002661398","display_name":"Dafeng Zhang","orcid":null},"institutions":[{"id":"https://openalex.org/I4210155230","display_name":"Samsung (China)","ror":"https://ror.org/04yt00889","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250650973","https://openalex.org/I4210155230"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Dafeng Zhang","raw_affiliation_strings":["Samsung Research China - Beijing (SRC-B)"],"affiliations":[{"raw_affiliation_string":"Samsung Research China - Beijing (SRC-B)","institution_ids":["https://openalex.org/I4210155230"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100352229","display_name":"Xin Liu","orcid":"https://orcid.org/0000-0002-4611-9790"},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xin Liu","raw_affiliation_strings":["Tsinghua University,Department of Electronic Engineering,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Electronic Engineering,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5052422844","display_name":"Yali Li","orcid":null},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yali Li","raw_affiliation_strings":["Tsinghua University,Department of Electronic Engineering,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Electronic Engineering,China","institution_ids":["https://openalex.org/I99065089"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5113052507","display_name":"Shengjin Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I99065089","display_name":"Tsinghua University","ror":"https://ror.org/03cve4549","country_code":"CN","type":"education","lineage":["https://openalex.org/I99065089"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shengjin Wang","raw_affiliation_strings":["Tsinghua University,Department of Electronic Engineering,China"],"affiliations":[{"raw_affiliation_string":"Tsinghua University,Department of Electronic Engineering,China","institution_ids":["https://openalex.org/I99065089"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":6,"corresponding_author_ids":["https://openalex.org/A5068996540"],"corresponding_institution_ids":["https://openalex.org/I99065089"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.04096133,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"5"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11439","display_name":"Video Analysis and Summarization","score":0.9065999984741211,"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/T11439","display_name":"Video Analysis and Summarization","score":0.9065999984741211,"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/computer-science","display_name":"Computer science","score":0.7664222121238708},{"id":"https://openalex.org/keywords/tower","display_name":"Tower","score":0.6797365546226501},{"id":"https://openalex.org/keywords/multimedia","display_name":"Multimedia","score":0.49015530943870544},{"id":"https://openalex.org/keywords/computer-graphics","display_name":"Computer graphics (images)","score":0.43463751673698425},{"id":"https://openalex.org/keywords/information-retrieval","display_name":"Information retrieval","score":0.36292535066604614},{"id":"https://openalex.org/keywords/computer-vision","display_name":"Computer vision","score":0.32556432485580444},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3207668662071228},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.0991847813129425}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7664222121238708},{"id":"https://openalex.org/C2777831296","wikidata":"https://www.wikidata.org/wiki/Q12518","display_name":"Tower","level":2,"score":0.6797365546226501},{"id":"https://openalex.org/C49774154","wikidata":"https://www.wikidata.org/wiki/Q131765","display_name":"Multimedia","level":1,"score":0.49015530943870544},{"id":"https://openalex.org/C121684516","wikidata":"https://www.wikidata.org/wiki/Q7600677","display_name":"Computer graphics (images)","level":1,"score":0.43463751673698425},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.36292535066604614},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.32556432485580444},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3207668662071228},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.0991847813129425},{"id":"https://openalex.org/C147176958","wikidata":"https://www.wikidata.org/wiki/Q77590","display_name":"Civil engineering","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp49660.2025.10890678","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp49660.2025.10890678","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2025 - 2025 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":22,"referenced_works":["https://openalex.org/W2425121537","https://openalex.org/W2560647685","https://openalex.org/W2765716052","https://openalex.org/W2963916161","https://openalex.org/W2984008963","https://openalex.org/W2989322838","https://openalex.org/W3121523901","https://openalex.org/W3175961224","https://openalex.org/W3204588463","https://openalex.org/W4309674289","https://openalex.org/W4386071687","https://openalex.org/W4390873732","https://openalex.org/W4390889724","https://openalex.org/W4402670856","https://openalex.org/W4402671548","https://openalex.org/W4402727142","https://openalex.org/W4402727764","https://openalex.org/W4402996663","https://openalex.org/W4404781975","https://openalex.org/W6853116092","https://openalex.org/W6856227841","https://openalex.org/W6858023062"],"related_works":["https://openalex.org/W4214759293","https://openalex.org/W2948890638","https://openalex.org/W2377743247","https://openalex.org/W2481514411","https://openalex.org/W2567029817","https://openalex.org/W2367420223","https://openalex.org/W2319168490","https://openalex.org/W2549137639","https://openalex.org/W3112969627","https://openalex.org/W3213682227"],"abstract_inverted_index":{"The":[0,128],"Large":[1],"Vision-Language":[2],"Model":[3],"(LVLM)":[4],"has":[5],"achieved":[6,182],"impressive":[7],"performance":[8],"in":[9,125,177],"the":[10,27,61,66,71,75,79,90,93,132,161,178,194],"field":[11],"of":[12,32,56,74,92,101,130,163,197],"visual-language":[13],"understanding.":[14],"However,":[15],"its":[16],"ability":[17],"to":[18,26,65,83,110,139],"understand":[19],"longer":[20],"videos":[21,41,77,115,176],"is":[22],"still":[23],"limited":[24],"due":[25],"length":[28],"and":[29,69,118,145,147,181,190],"information":[30],"diversity":[31],"multi-modal":[33,174],"videos.":[34],"Moreover,":[35],"accurately":[36],"matching":[37],"detailed":[38],"content":[39],"within":[40,89],"remains":[42],"an":[43],"open":[44,179],"research":[45],"problem.":[46],"We":[47],"design":[48],"a":[49,99],"new":[50,188],"framework":[51],"for":[52,86,193],"LVLM":[53],"inference,":[54],"\"Tower":[55],"Thoughts\"":[57],"(ToT),":[58],"which":[59,106],"extends":[60],"\"Chain-of-Thought\"":[62],"(CoT)":[63],"approach":[64],"visual":[67,198],"domain":[68],"constructs":[70],"high-dimensional":[72],"semantics":[73],"complete":[76],"from":[78,113],"bottom":[80],"up.":[81],"Meanwhile,":[82],"achieve":[84],"question-answering":[85],"video":[87,120,154],"details":[88,112],"constraints":[91],"restricted":[94],"context":[95],"window,":[96],"we":[97,169],"propose":[98],"method":[100],"self-retrieval":[102],"augmented":[103],"generation":[104,149],"(SRAG),":[105],"makes":[107],"it":[108],"possible":[109],"obtain":[111],"long":[114,175],"by":[116],"storing":[117],"accessing":[119],"text":[121],"as":[122],"dense":[123],"vectors":[124],"non-parametric":[126],"memory.":[127],"solution":[129],"combining":[131],"ToT":[133],"with":[134],"SRAG":[135],"enables":[136],"our":[137,164],"model":[138],"have":[140],"cross-modal":[141],"high-density":[142],"semantic":[143],"fusion":[144],"comprehensive":[146],"accurate":[148],"capabilities,":[150],"thereby":[151],"achieving":[152],"rationalized":[153],"answers.":[155],"Experiments":[156],"on":[157,173],"public":[158],"benchmarks":[159],"demonstrate":[160],"effectiveness":[162],"proposed":[165],"method.":[166],"In":[167],"addition,":[168],"also":[170],"conducted":[171],"experiments":[172],"world":[180],"remarkable":[183],"outcomes.":[184],"These":[185],"results":[186],"provide":[187],"perspectives":[189],"technical":[191],"routes":[192],"future":[195],"development":[196],"language":[199],"models.":[200]},"counts_by_year":[],"updated_date":"2025-12-21T01:58:51.020947","created_date":"2025-10-10T00:00:00"}