{"id":"https://openalex.org/W4392904379","doi":"https://doi.org/10.1109/icassp48485.2024.10447954","title":"Multimodal Transformer with a Low-Computational-Cost Guarantee","display_name":"Multimodal Transformer with a Low-Computational-Cost Guarantee","publication_year":2024,"publication_date":"2024-03-18","ids":{"openalex":"https://openalex.org/W4392904379","doi":"https://doi.org/10.1109/icassp48485.2024.10447954"},"language":"en","primary_location":{"id":"doi:10.1109/icassp48485.2024.10447954","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447954","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 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/A5027447596","display_name":"Sungjin Park","orcid":"https://orcid.org/0000-0002-1447-4536"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":true,"raw_author_name":"Sungjin Park","raw_affiliation_strings":["KAIST"],"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I4210099236","https://openalex.org/I157485424"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5034622258","display_name":"Edward Choi","orcid":"https://orcid.org/0000-0002-5958-3509"},"institutions":[{"id":"https://openalex.org/I157485424","display_name":"Korea Advanced Institute of Science and Technology","ror":"https://ror.org/05apxxy63","country_code":"KR","type":"education","lineage":["https://openalex.org/I157485424"]},{"id":"https://openalex.org/I4210099236","display_name":"Kootenay Association for Science & Technology","ror":"https://ror.org/011pv9p44","country_code":"CA","type":"nonprofit","lineage":["https://openalex.org/I4210099236"]}],"countries":["CA","KR"],"is_corresponding":false,"raw_author_name":"Edward Choi","raw_affiliation_strings":["KAIST"],"affiliations":[{"raw_affiliation_string":"KAIST","institution_ids":["https://openalex.org/I4210099236","https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":2,"corresponding_author_ids":["https://openalex.org/A5027447596"],"corresponding_institution_ids":["https://openalex.org/I157485424","https://openalex.org/I4210099236"],"apc_list":null,"apc_paid":null,"fwci":0.735,"has_fulltext":false,"cited_by_count":3,"citation_normalized_percentile":{"value":0.68972421,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"6125","last_page":"6129"},"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/T10812","display_name":"Human Pose and Action Recognition","score":0.9994000196456909,"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":0.9962000250816345,"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/computer-science","display_name":"Computer science","score":0.7867274880409241},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.7497227787971497},{"id":"https://openalex.org/keywords/action-recognition","display_name":"Action recognition","score":0.5960003733634949},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5666731595993042},{"id":"https://openalex.org/keywords/flops","display_name":"FLOPS","score":0.5521730184555054},{"id":"https://openalex.org/keywords/modalities","display_name":"Modalities","score":0.5229057669639587},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.48664435744285583},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.4516059160232544},{"id":"https://openalex.org/keywords/computational-complexity-theory","display_name":"Computational complexity theory","score":0.43906304240226746},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.14044272899627686},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.08383139967918396},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.06872925162315369}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7867274880409241},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.7497227787971497},{"id":"https://openalex.org/C2987834672","wikidata":"https://www.wikidata.org/wiki/Q4677630","display_name":"Action recognition","level":3,"score":0.5960003733634949},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5666731595993042},{"id":"https://openalex.org/C3826847","wikidata":"https://www.wikidata.org/wiki/Q188768","display_name":"FLOPS","level":2,"score":0.5521730184555054},{"id":"https://openalex.org/C2779903281","wikidata":"https://www.wikidata.org/wiki/Q6888026","display_name":"Modalities","level":2,"score":0.5229057669639587},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.48664435744285583},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.4516059160232544},{"id":"https://openalex.org/C179799912","wikidata":"https://www.wikidata.org/wiki/Q205084","display_name":"Computational complexity theory","level":2,"score":0.43906304240226746},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.14044272899627686},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.08383139967918396},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.06872925162315369},{"id":"https://openalex.org/C36289849","wikidata":"https://www.wikidata.org/wiki/Q34749","display_name":"Social science","level":1,"score":0.0},{"id":"https://openalex.org/C144024400","wikidata":"https://www.wikidata.org/wiki/Q21201","display_name":"Sociology","level":0,"score":0.0},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0},{"id":"https://openalex.org/C2777212361","wikidata":"https://www.wikidata.org/wiki/Q5127848","display_name":"Class (philosophy)","level":2,"score":0.0},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icassp48485.2024.10447954","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icassp48485.2024.10447954","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ICASSP 2024 - 2024 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":32,"referenced_works":["https://openalex.org/W2053101950","https://openalex.org/W2560730294","https://openalex.org/W2593116425","https://openalex.org/W2619947201","https://openalex.org/W2962960500","https://openalex.org/W2963890019","https://openalex.org/W2970231061","https://openalex.org/W3015468748","https://openalex.org/W3033529678","https://openalex.org/W3106298483","https://openalex.org/W3131922516","https://openalex.org/W3196974791","https://openalex.org/W3207758636","https://openalex.org/W4214612132","https://openalex.org/W4323654151","https://openalex.org/W4360600216","https://openalex.org/W4385245566","https://openalex.org/W4386071687","https://openalex.org/W4390874283","https://openalex.org/W6753640285","https://openalex.org/W6755207826","https://openalex.org/W6766904570","https://openalex.org/W6771626834","https://openalex.org/W6781533629","https://openalex.org/W6783944145","https://openalex.org/W6784333009","https://openalex.org/W6789753369","https://openalex.org/W6790830454","https://openalex.org/W6797613833","https://openalex.org/W6802745747","https://openalex.org/W6839015040","https://openalex.org/W6842585177"],"related_works":["https://openalex.org/W4315697128","https://openalex.org/W3102845713","https://openalex.org/W2971502891","https://openalex.org/W3205506801","https://openalex.org/W4280599700","https://openalex.org/W2185469136","https://openalex.org/W3183570023","https://openalex.org/W4382323155","https://openalex.org/W2016508734","https://openalex.org/W4283332100"],"abstract_inverted_index":{"Transformer-based":[0],"models":[1],"have":[2],"significantly":[3,29],"improved":[4],"performance":[5,77],"across":[6],"a":[7,32,60,99],"range":[8],"of":[9,35,48],"multimodal":[10,27,62,83,94,106,113],"understanding":[11],"tasks,":[12],"such":[13],"as":[14,45],"visual":[15],"question":[16],"answering":[17],"[1],":[18],"[2]":[19],"and":[20,73,96,117],"action":[21],"recognition":[22],"[3],":[23],"[4],":[24],"[5].":[25],"However,":[26],"Transformers":[28],"suffer":[30],"from":[31],"quadratic":[33],"complexity":[34],"the":[36,40,46],"multi-head":[37],"attention":[38,63,84,88],"with":[39,75],"input":[41],"sequence":[42],"length,":[43],"especially":[44],"number":[47],"modalities":[49],"increases.":[50],"To":[51],"address":[52],"this,":[53],"we":[54],"introduce":[55],"Low-Cost":[56],"Multimodal":[57],"Transformer":[58,107],"(LoCoMT),":[59],"novel":[61],"mechanism":[64],"that":[65,120],"aims":[66],"to":[67,86,104],"reduce":[68],"computational":[69,101],"cost":[70,102],"during":[71],"training":[72],"inference":[74],"minimal":[76],"loss.":[78],"Specifically,":[79],"by":[80],"assigning":[81],"different":[82],"patterns":[85],"each":[87],"head,":[89],"LoCoMT":[90,121],"can":[91],"flexibly":[92],"control":[93],"signals":[95],"theoretically":[97],"ensures":[98],"reduced":[100],"compared":[103],"existing":[105],"variants.":[108],"Experimental":[109],"results":[110],"on":[111],"two":[112],"datasets,":[114],"namely":[115],"Audioset":[116],"MedVidCL":[118],"demonstrate":[119],"not":[122],"only":[123],"reduces":[124],"GFLOPs":[125],"but":[126],"also":[127],"matches":[128],"or":[129],"even":[130],"outperforms":[131],"established":[132],"models.":[133]},"counts_by_year":[{"year":2026,"cited_by_count":1},{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}