{"id":"https://openalex.org/W4415538176","doi":"https://doi.org/10.1145/3746027.3755792","title":"VISA: Group-wise Visual Token Selection and Aggregation via Graph Summarization for Efficient MLLMs Inference","display_name":"VISA: Group-wise Visual Token Selection and Aggregation via Graph Summarization for Efficient MLLMs Inference","publication_year":2025,"publication_date":"2025-10-25","ids":{"openalex":"https://openalex.org/W4415538176","doi":"https://doi.org/10.1145/3746027.3755792"},"language":"en","primary_location":{"id":"doi:10.1145/3746027.3755792","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3755792","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"green","oa_url":"https://arxiv.org/pdf/2508.17857","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Pengfei Jiang","orcid":"https://orcid.org/0009-0001-3969-5490"},"institutions":[{"id":"https://openalex.org/I191208505","display_name":"Xiamen University","ror":"https://ror.org/00mcjh785","country_code":"CN","type":"education","lineage":["https://openalex.org/I191208505"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Pengfei Jiang","raw_affiliation_strings":["Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China"],"raw_orcid":"https://orcid.org/0009-0001-3969-5490","affiliations":[{"raw_affiliation_string":"Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China","institution_ids":["https://openalex.org/I191208505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5056070516","display_name":"Hanjun Li","orcid":null},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hanjun Li","raw_affiliation_strings":["Tencent Youtu Lab, Shenzhen, China"],"raw_orcid":"https://orcid.org/0009-0006-4211-7479","affiliations":[{"raw_affiliation_string":"Tencent Youtu Lab, Shenzhen, China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Linglan Zhao","orcid":"https://orcid.org/0000-0002-2241-6977"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Linglan Zhao","raw_affiliation_strings":["Tencent Youtu Lab, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0002-2241-6977","affiliations":[{"raw_affiliation_string":"Tencent Youtu Lab, Shanghai, China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5000389309","display_name":"Fei Chao","orcid":"https://orcid.org/0000-0002-6928-2638"},"institutions":[{"id":"https://openalex.org/I191208505","display_name":"Xiamen University","ror":"https://ror.org/00mcjh785","country_code":"CN","type":"education","lineage":["https://openalex.org/I191208505"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Fei Chao","raw_affiliation_strings":["Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China"],"raw_orcid":"https://orcid.org/0000-0002-6928-2638","affiliations":[{"raw_affiliation_string":"Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China","institution_ids":["https://openalex.org/I191208505"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101967246","display_name":"Ke Yan","orcid":"https://orcid.org/0000-0003-3424-4866"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ke Yan","raw_affiliation_strings":["Tencent Youtu Lab, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0003-3424-4866","affiliations":[{"raw_affiliation_string":"Tencent Youtu Lab, Shanghai, China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5086397952","display_name":"Shouhong Ding","orcid":"https://orcid.org/0000-0002-3175-3553"},"institutions":[{"id":"https://openalex.org/I2250653659","display_name":"Tencent (China)","ror":"https://ror.org/00hhjss72","country_code":"CN","type":"company","lineage":["https://openalex.org/I2250653659"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shouhong Ding","raw_affiliation_strings":["Tencent Youtu Lab, Shanghai, China"],"raw_orcid":"https://orcid.org/0000-0002-3175-3553","affiliations":[{"raw_affiliation_string":"Tencent Youtu Lab, Shanghai, China","institution_ids":["https://openalex.org/I2250653659"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5016080094","display_name":"Rongrong Ji","orcid":"https://orcid.org/0000-0001-9163-2932"},"institutions":[{"id":"https://openalex.org/I191208505","display_name":"Xiamen University","ror":"https://ror.org/00mcjh785","country_code":"CN","type":"education","lineage":["https://openalex.org/I191208505"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Rongrong Ji","raw_affiliation_strings":["Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China"],"raw_orcid":"https://orcid.org/0000-0001-9163-2932","affiliations":[{"raw_affiliation_string":"Key Laboratory of Multimedia Trusted Perception and Efficient Computing, Ministry of Education of China, Xiamen University, Xiamen, China","institution_ids":["https://openalex.org/I191208505"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I191208505"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.28726169,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"11130","last_page":"11139"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11714","display_name":"Multimodal Machine Learning Applications","score":0.9998999834060669,"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":0.9998999834060669,"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/T11307","display_name":"Domain Adaptation and Few-Shot Learning","score":0.998199999332428,"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"}},{"id":"https://openalex.org/T10812","display_name":"Human Pose and Action Recognition","score":0.9954000115394592,"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":0.8136000037193298},{"id":"https://openalex.org/keywords/automatic-summarization","display_name":"Automatic summarization","score":0.6348000168800354},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5928000211715698},{"id":"https://openalex.org/keywords/visualization","display_name":"Visualization","score":0.5526000261306763},{"id":"https://openalex.org/keywords/selection","display_name":"Selection (genetic algorithm)","score":0.5425999760627747},{"id":"https://openalex.org/keywords/graph","display_name":"Graph","score":0.5410000085830688},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.39739999175071716}],"concepts":[{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.8136000037193298},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7998999953269958},{"id":"https://openalex.org/C170858558","wikidata":"https://www.wikidata.org/wiki/Q1394144","display_name":"Automatic summarization","level":2,"score":0.6348000168800354},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5928000211715698},{"id":"https://openalex.org/C36464697","wikidata":"https://www.wikidata.org/wiki/Q451553","display_name":"Visualization","level":2,"score":0.5526000261306763},{"id":"https://openalex.org/C81917197","wikidata":"https://www.wikidata.org/wiki/Q628760","display_name":"Selection (genetic algorithm)","level":2,"score":0.5425999760627747},{"id":"https://openalex.org/C132525143","wikidata":"https://www.wikidata.org/wiki/Q141488","display_name":"Graph","level":2,"score":0.5410000085830688},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.5303999781608582},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.39739999175071716},{"id":"https://openalex.org/C124304363","wikidata":"https://www.wikidata.org/wiki/Q673661","display_name":"Abstraction","level":2,"score":0.32820001244544983},{"id":"https://openalex.org/C153180895","wikidata":"https://www.wikidata.org/wiki/Q7148389","display_name":"Pattern recognition (psychology)","level":2,"score":0.3269999921321869},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.3156000077724457},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3122999966144562},{"id":"https://openalex.org/C178253425","wikidata":"https://www.wikidata.org/wiki/Q162668","display_name":"Visual perception","level":3,"score":0.29660001397132874},{"id":"https://openalex.org/C103278499","wikidata":"https://www.wikidata.org/wiki/Q254465","display_name":"Similarity (geometry)","level":3,"score":0.2946999967098236},{"id":"https://openalex.org/C2779668609","wikidata":"https://www.wikidata.org/wiki/Q623092","display_name":"Rapid serial visual presentation","level":3,"score":0.28999999165534973},{"id":"https://openalex.org/C23123220","wikidata":"https://www.wikidata.org/wiki/Q816826","display_name":"Information retrieval","level":1,"score":0.2881999909877777},{"id":"https://openalex.org/C52622490","wikidata":"https://www.wikidata.org/wiki/Q1026626","display_name":"Feature extraction","level":2,"score":0.2770000100135803},{"id":"https://openalex.org/C59732488","wikidata":"https://www.wikidata.org/wiki/Q2528440","display_name":"Visual analytics","level":3,"score":0.26840001344680786},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.26840001344680786},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.26809999346733093},{"id":"https://openalex.org/C130318100","wikidata":"https://www.wikidata.org/wiki/Q2268914","display_name":"Semantic similarity","level":2,"score":0.2597000002861023}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3746027.3755792","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3746027.3755792","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 33rd ACM International Conference on Multimedia","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2508.17857","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.17857","pdf_url":"https://arxiv.org/pdf/2508.17857","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"}],"best_oa_location":{"id":"pmh:oai:arXiv.org:2508.17857","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2508.17857","pdf_url":"https://arxiv.org/pdf/2508.17857","source":{"id":"https://openalex.org/S4393918464","display_name":"ArXiv.org","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G1656144382","display_name":null,"funder_award_id":"No.62025603","funder_id":"https://openalex.org/F4320336125","funder_display_name":"National Science Fund for Distinguished Young Scholars"},{"id":"https://openalex.org/G5348289504","display_name":null,"funder_award_id":"No.U21B2037,No.U22B2051,No.U23A20383,No.U21A20472,No.62176222,No.62176223,No.62176226,No.62072386,No.62072387,No.62072389,No.62002305,No.62272401","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320336125","display_name":"National Science Fund for Distinguished Young Scholars","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":5,"referenced_works":["https://openalex.org/W2560730294","https://openalex.org/W2963518342","https://openalex.org/W2963622213","https://openalex.org/W2964220823","https://openalex.org/W4402727764"],"related_works":[],"abstract_inverted_index":{"In":[0],"this":[1,92],"study,":[2],"we":[3,102],"introduce":[4,103],"a":[5,55,68,71,95,104,170],"novel":[6],"method":[7,42,164],"called":[8],"group-wise":[9,105],"VI":[10],"sual":[11],"token":[12,38,58,66,99,106],"Selection":[13],"and":[14,116,152,176],"Aggregation":[15],"(VISA)":[16],"to":[17,110,157],"address":[18],"the":[19,124,137,140,159],"issue":[20],"of":[21,127,139,161],"inefficient":[22],"inference":[23,177],"stemming":[24],"from":[25,84,123],"excessive":[26],"visual":[27,46,50,57,65,78,98,112,134,141],"tokens":[28,86,89,113,122],"in":[29],"multimoal":[30],"large":[31],"language":[32],"models":[33],"(MLLMs).":[34],"Compared":[35],"with":[36],"previous":[37,167],"pruning":[39],"approaches,":[40],"our":[41],"can":[43],"preserve":[44],"more":[45,96],"information":[47,83,142],"while":[48],"compressing":[49],"tokens.":[51,79],"We":[52,145],"first":[53],"propose":[54],"graph-based":[56],"aggregation":[59],"(VTA)":[60],"module.":[61],"VTA":[62],"treats":[63],"each":[64,128],"as":[67],"node,":[69],"forming":[70],"graph":[72],"based":[73,90],"on":[74,91,149],"semantic":[75],"similarity":[76],"among":[77],"It":[80],"then":[81],"aggregates":[82,133],"removed":[85,117],"into":[87,114],"kept":[88,115],"graph,":[93],"producing":[94],"compact":[97],"representation.":[100],"Additionally,":[101],"selection":[107],"strategy":[108,131],"(GTS)":[109],"divide":[111],"ones,":[118],"guided":[119],"by":[120],"text":[121],"final":[125],"layers":[126],"group.":[129],"This":[130],"progressively":[132],"information,":[135],"enhancing":[136],"stability":[138],"extraction":[143],"process.":[144],"conduct":[146],"comprehensive":[147],"experiments":[148],"LLaVA-1.5,":[150],"LLaVA-NeXT,":[151],"Video-LLaVA":[153],"across":[154],"various":[155],"benchmarks":[156],"validate":[158],"efficacy":[160],"VISA.":[162],"Our":[163],"consistently":[165],"outperforms":[166],"methods,":[168],"achieving":[169],"superior":[171],"trade-off":[172],"between":[173],"model":[174],"performance":[175],"speed.":[178]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-25T00:00:00"}