{"id":"https://openalex.org/W4414360630","doi":"https://doi.org/10.24963/ijcai.2025/258","title":"DToMA: Training-free Dynamic Token MAnipulation for Long Video Understanding","display_name":"DToMA: Training-free Dynamic Token MAnipulation for Long Video Understanding","publication_year":2025,"publication_date":"2025-09-01","ids":{"openalex":"https://openalex.org/W4414360630","doi":"https://doi.org/10.24963/ijcai.2025/258"},"language":"en","primary_location":{"id":"doi:10.24963/ijcai.2025/258","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/258","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"},"type":"conference-paper","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/A5071874503","display_name":"Bowen Yuan","orcid":"https://orcid.org/0000-0002-3012-4077"},"institutions":[{"id":"https://openalex.org/I41198531","display_name":"Nanjing University of Posts and Telecommunications","ror":"https://ror.org/043bpky34","country_code":"CN","type":"education","lineage":["https://openalex.org/I41198531"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bowen Yuan","raw_affiliation_strings":["Nanjing University of Posts and Telecommunications"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanjing University of Posts and Telecommunications","institution_ids":["https://openalex.org/I41198531"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101970691","display_name":"Sisi You","orcid":"https://orcid.org/0009-0004-2173-1694"},"institutions":[{"id":"https://openalex.org/I41198531","display_name":"Nanjing University of Posts and Telecommunications","ror":"https://ror.org/043bpky34","country_code":"CN","type":"education","lineage":["https://openalex.org/I41198531"]},{"id":"https://openalex.org/I4210159812","display_name":"Tibetan Traditional Medical College","ror":"https://ror.org/05hk7qq56","country_code":"CN","type":"education","lineage":["https://openalex.org/I4210159812"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Sisi You","raw_affiliation_strings":["Nanjing University of Posts and Telecommunications","State Key Laboratory of Tibetan Intelligence"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanjing University of Posts and Telecommunications","institution_ids":["https://openalex.org/I41198531"]},{"raw_affiliation_string":"State Key Laboratory of Tibetan Intelligence","institution_ids":["https://openalex.org/I4210159812"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5007962086","display_name":"Bing\u2010Kun Bao","orcid":"https://orcid.org/0000-0001-5956-831X"},"institutions":[{"id":"https://openalex.org/I41198531","display_name":"Nanjing University of Posts and Telecommunications","ror":"https://ror.org/043bpky34","country_code":"CN","type":"education","lineage":["https://openalex.org/I41198531"]},{"id":"https://openalex.org/I4210136793","display_name":"Peng Cheng Laboratory","ror":"https://ror.org/03qdqbt06","country_code":"CN","type":"facility","lineage":["https://openalex.org/I4210136793"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bing-Kun Bao","raw_affiliation_strings":["Nanjing University of Posts and Telecommunications","Pengcheng Laboratory"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Nanjing University of Posts and Telecommunications","institution_ids":["https://openalex.org/I41198531"]},{"raw_affiliation_string":"Pengcheng Laboratory","institution_ids":["https://openalex.org/I4210136793"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"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":"2314","last_page":"2322"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T13114","display_name":"Image Processing Techniques and Applications","score":0.9948999881744385,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T13114","display_name":"Image Processing Techniques and Applications","score":0.9948999881744385,"subfield":{"id":"https://openalex.org/subfields/2214","display_name":"Media Technology"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T12357","display_name":"Digital Media Forensic Detection","score":0.9907000064849854,"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/T12111","display_name":"Industrial Vision Systems and Defect Detection","score":0.9537000060081482,"subfield":{"id":"https://openalex.org/subfields/2209","display_name":"Industrial and Manufacturing Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.6367999911308289},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.5483999848365784},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.45980000495910645},{"id":"https://openalex.org/keywords/memorization","display_name":"Memorization","score":0.4424000084400177},{"id":"https://openalex.org/keywords/representation","display_name":"Representation (politics)","score":0.4214000105857849},{"id":"https://openalex.org/keywords/flexibility","display_name":"Flexibility (engineering)","score":0.3880000114440918},{"id":"https://openalex.org/keywords/human-visual-system-model","display_name":"Human visual system model","score":0.36309999227523804},{"id":"https://openalex.org/keywords/reduction","display_name":"Reduction (mathematics)","score":0.3522999882698059},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.3483000099658966}],"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":0.6367999911308289},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5663999915122986},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.5483999848365784},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.45980000495910645},{"id":"https://openalex.org/C30038468","wikidata":"https://www.wikidata.org/wiki/Q4354775","display_name":"Memorization","level":2,"score":0.4424000084400177},{"id":"https://openalex.org/C2776359362","wikidata":"https://www.wikidata.org/wiki/Q2145286","display_name":"Representation (politics)","level":3,"score":0.4214000105857849},{"id":"https://openalex.org/C2780598303","wikidata":"https://www.wikidata.org/wiki/Q65921492","display_name":"Flexibility (engineering)","level":2,"score":0.3880000114440918},{"id":"https://openalex.org/C160086991","wikidata":"https://www.wikidata.org/wiki/Q5939193","display_name":"Human visual system model","level":3,"score":0.36309999227523804},{"id":"https://openalex.org/C111335779","wikidata":"https://www.wikidata.org/wiki/Q3454686","display_name":"Reduction (mathematics)","level":2,"score":0.3522999882698059},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.3483000099658966},{"id":"https://openalex.org/C169900460","wikidata":"https://www.wikidata.org/wiki/Q2200417","display_name":"Cognition","level":2,"score":0.34709998965263367},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.34380000829696655},{"id":"https://openalex.org/C203357204","wikidata":"https://www.wikidata.org/wiki/Q1089605","display_name":"Chunking (psychology)","level":2,"score":0.34200000762939453},{"id":"https://openalex.org/C107457646","wikidata":"https://www.wikidata.org/wiki/Q207434","display_name":"Human\u2013computer interaction","level":1,"score":0.326200008392334},{"id":"https://openalex.org/C66024118","wikidata":"https://www.wikidata.org/wiki/Q1122506","display_name":"Computational model","level":2,"score":0.32339999079704285},{"id":"https://openalex.org/C511192102","wikidata":"https://www.wikidata.org/wiki/Q5156948","display_name":"Comprehension","level":2,"score":0.31869998574256897},{"id":"https://openalex.org/C100776233","wikidata":"https://www.wikidata.org/wiki/Q2532492","display_name":"Bridge (graph theory)","level":2,"score":0.31299999356269836},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2856999933719635},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.28459998965263367},{"id":"https://openalex.org/C2776760102","wikidata":"https://www.wikidata.org/wiki/Q5139990","display_name":"Code (set theory)","level":3,"score":0.27869999408721924},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.2732999920845032},{"id":"https://openalex.org/C94124525","wikidata":"https://www.wikidata.org/wiki/Q912550","display_name":"Categorization","level":2,"score":0.2678999900817871},{"id":"https://openalex.org/C108583219","wikidata":"https://www.wikidata.org/wiki/Q197536","display_name":"Deep learning","level":2,"score":0.2644999921321869},{"id":"https://openalex.org/C97931131","wikidata":"https://www.wikidata.org/wiki/Q5282087","display_name":"Discriminative model","level":2,"score":0.2572999894618988},{"id":"https://openalex.org/C2778251979","wikidata":"https://www.wikidata.org/wiki/Q7936617","display_name":"Visual processing","level":3,"score":0.2563000023365021}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.24963/ijcai.2025/258","is_oa":false,"landing_page_url":"https://doi.org/10.24963/ijcai.2025/258","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the Thirty-Fourth International Joint Conference on Artificial Intelligence","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Video":[0],"Large":[1],"Language":[2],"Models":[3],"(VideoLLMs)":[4],"often":[5,37],"require":[6],"thousands":[7],"of":[8,49],"visual":[9,23,145,157],"tokens":[10,158],"to":[11,16,81,95,137,147,155],"process":[12],"long":[13,164],"videos,":[14],"leading":[15],"substantial":[17],"computational":[18],"costs,":[19],"further":[20],"exacerbated":[21],"by":[22,126],"token":[24,27],"inefficiency.":[25],"Existing":[26],"reduction":[28],"and":[29,60,100,140,151,174,179,186],"alternative":[30],"video":[31,84,107,165],"representation":[32],"methods":[33,178],"improve":[34],"efficiency":[35,173],"but":[36],"compromise":[38],"comprehension":[39],"abilities.":[40],"In":[41],"this":[42,114],"work,":[43],"we":[44,116],"analyze":[45],"the":[46],"reasoning":[47,57],"processes":[48],"VideoLLMs":[50,79],"in":[51,76,90,101,130],"multi-choice":[52],"VideoQA":[53],"task,":[54],"identifying":[55],"three":[56,131],"stages\u2014shallow,":[58],"intermediate,":[59],"deep":[61,102],"stages\u2014that":[62],"closely":[63],"mimic":[64],"human":[65,127],"cognitive":[66],"processing.":[67],"Our":[68],"analysis":[69],"reveals":[70],"specific":[71],"inefficiencies":[72],"at":[73,191],"each":[74],"stage:":[75],"shallow":[77],"layers,":[78,92,103],"attempt":[80],"memorize":[82],"all":[83],"details":[85],"without":[86],"prioritizing":[87],"relevant":[88],"content;":[89],"intermediate":[91],"models":[93],"fail":[94],"re-examine":[96],"uncertain":[97],"content":[98,149],"dynamically;":[99],"they":[104],"continue":[105],"processing":[106],"even":[108],"when":[109,159],"sufficiently":[110],"confident.":[111,160],"To":[112],"bridge":[113],"gap,":[115],"propose":[117],"DToMA,":[118],"a":[119],"training-free":[120],"Dynamic":[121],"Token":[122],"MAnipulation":[123],"method":[124],"inspired":[125],"adjustment":[128],"mechanisms":[129],"aspects:":[132],"1)":[133],"Text-guided":[134],"keyframe-aware":[135],"reorganization":[136],"prioritize":[138],"keyframes":[139],"reduce":[141],"redundancy,":[142],"2)":[143],"Uncertainty-based":[144],"injection":[146],"revisit":[148],"dynamically,":[150],"3)":[152],"Early-exit":[153],"pruning":[154],"halt":[156],"Experiments":[161],"on":[162],"6":[163],"understanding":[166],"benchmarks":[167],"show":[168],"that":[169],"DToMA":[170],"enhances":[171],"both":[172],"comprehension,":[175],"outperforming":[176],"state-of-the-art":[177],"generalizing":[180],"well":[181],"across":[182],"3":[183],"VideoLLM":[184],"architectures":[185],"sizes.":[187],"Code":[188],"is":[189],"available":[190],"https://github.com/yuanrr/DToMA.":[192]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-10-10T00:00:00"}