{"id":"https://openalex.org/W7155502461","doi":"https://doi.org/10.1145/3767295.3769389","title":"Efficient Multimodal Serving via Module Multiplexing","display_name":"Efficient Multimodal Serving via Module Multiplexing","publication_year":2026,"publication_date":"2026-04-24","ids":{"openalex":"https://openalex.org/W7155502461","doi":"https://doi.org/10.1145/3767295.3769389"},"language":"en","primary_location":{"id":"doi:10.1145/3767295.3769389","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3769389","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st European Conference on Computer Systems","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3767295.3769389","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5075795671","display_name":"Zicong Hong","orcid":"https://orcid.org/0000-0001-5689-382X"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Zicong Hong","raw_affiliation_strings":["Hong Kong University of Science and Technology, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0001-5689-382X","affiliations":[{"raw_affiliation_string":"Hong Kong University of Science and Technology, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009453810","display_name":"Yuyan Chen","orcid":"https://orcid.org/0000-0003-1771-0837"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yuyan Chen","raw_affiliation_strings":["Sun Yat-sen University, Zhuhai, China"],"raw_orcid":"https://orcid.org/0009-0002-7738-640X","affiliations":[{"raw_affiliation_string":"Sun Yat-sen University, Zhuhai, China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5105398187","display_name":"H ZHANG","orcid":null},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Haoyue Zhang","raw_affiliation_strings":["Hong Kong University of Science and Technology, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0009-0002-8308-3827","affiliations":[{"raw_affiliation_string":"Hong Kong University of Science and Technology, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066098631","display_name":"Peng Li","orcid":"https://orcid.org/0000-0002-5303-0700"},"institutions":[{"id":"https://openalex.org/I87445476","display_name":"Xi'an Jiaotong University","ror":"https://ror.org/017zhmm22","country_code":"CN","type":"education","lineage":["https://openalex.org/I87445476"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Peng Li","raw_affiliation_strings":["Xi'an Jiaotong University, Xi'an, China"],"raw_orcid":"https://orcid.org/0000-0002-5303-0700","affiliations":[{"raw_affiliation_string":"Xi'an Jiaotong University, Xi'an, China","institution_ids":["https://openalex.org/I87445476"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5066482792","display_name":"Wuhui Chen","orcid":"https://orcid.org/0000-0003-4430-7904"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Wuhui Chen","raw_affiliation_strings":["Sun Yat-sen University, Zhuhai, China"],"raw_orcid":"https://orcid.org/0000-0003-4430-7904","affiliations":[{"raw_affiliation_string":"Sun Yat-sen University, Zhuhai, China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5043464306","display_name":"Song Guo","orcid":"https://orcid.org/0000-0001-9831-2202"},"institutions":[{"id":"https://openalex.org/I200769079","display_name":"Hong Kong University of Science and Technology","ror":"https://ror.org/00q4vv597","country_code":"HK","type":"education","lineage":["https://openalex.org/I200769079"]}],"countries":["HK"],"is_corresponding":false,"raw_author_name":"Song Guo","raw_affiliation_strings":["Hong Kong University of Science and Technology, Hong Kong, Hong Kong"],"raw_orcid":"https://orcid.org/0000-0001-9831-2202","affiliations":[{"raw_affiliation_string":"Hong Kong University of Science and Technology, Hong Kong, Hong Kong","institution_ids":["https://openalex.org/I200769079"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5134538408","display_name":"Xiaowei Shen","orcid":"https://orcid.org/0009-0002-5420-583X"},"institutions":[{"id":"https://openalex.org/I4210137016","display_name":"GlaxoSmithKline (China)","ror":"https://ror.org/03jrn4k58","country_code":"CN","type":"company","lineage":["https://openalex.org/I188950975","https://openalex.org/I4210137016"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaowei Shen","raw_affiliation_strings":["MetaX, Shanghai, China"],"raw_orcid":"https://orcid.org/0009-0002-5420-583X","affiliations":[{"raw_affiliation_string":"MetaX, Shanghai, China","institution_ids":["https://openalex.org/I4210137016"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":4,"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":"2292","last_page":"2307"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10860","display_name":"Speech and Audio Processing","score":0.19009999930858612,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T10860","display_name":"Speech and Audio Processing","score":0.19009999930858612,"subfield":{"id":"https://openalex.org/subfields/1711","display_name":"Signal Processing"},"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/T10500","display_name":"Sparse and Compressive Sensing Techniques","score":0.0803999975323677,"subfield":{"id":"https://openalex.org/subfields/2206","display_name":"Computational Mechanics"},"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/T10711","display_name":"Target Tracking and Data Fusion in Sensor Networks","score":0.03480000048875809,"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/multiplexing","display_name":"Multiplexing","score":0.4909000098705292},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.29170000553131104},{"id":"https://openalex.org/keywords/time-division-multiplexing","display_name":"Time-division multiplexing","score":0.27219998836517334},{"id":"https://openalex.org/keywords/encoding","display_name":"Encoding (memory)","score":0.265500009059906},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.2644999921321869}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.631600022315979},{"id":"https://openalex.org/C19275194","wikidata":"https://www.wikidata.org/wiki/Q222903","display_name":"Multiplexing","level":2,"score":0.4909000098705292},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.3490000069141388},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.3005000054836273},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.29170000553131104},{"id":"https://openalex.org/C50661577","wikidata":"https://www.wikidata.org/wiki/Q901831","display_name":"Time-division multiplexing","level":3,"score":0.27219998836517334},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.265500009059906},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2644999921321869},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.26440000534057617},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.25850000977516174},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2572999894618988},{"id":"https://openalex.org/C557945733","wikidata":"https://www.wikidata.org/wiki/Q389772","display_name":"Data transmission","level":2,"score":0.25540000200271606},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2540999948978424}],"mesh":[],"locations_count":2,"locations":[{"id":"doi:10.1145/3767295.3769389","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3769389","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st European Conference on Computer Systems","raw_type":"proceedings-article"},{"id":"pmh:oai:repository.hkust.edu.hk:1783.1-172777","is_oa":false,"landing_page_url":"http://repository.hkust.edu.hk/ir/Record/1783.1-172777","pdf_url":null,"source":{"id":"https://openalex.org/S4306401796","display_name":"Rare & Special e-Zone (The Hong Kong University of Science and Technology)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I200769079","host_organization_name":"Hong Kong University of Science and Technology","host_organization_lineage":["https://openalex.org/I200769079"],"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":"Conference paper"}],"best_oa_location":{"id":"doi:10.1145/3767295.3769389","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3767295.3769389","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 21st European Conference on Computer Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.45693641901016235,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":1,"referenced_works":["https://openalex.org/W4413120574"],"related_works":[],"abstract_inverted_index":{"Multimodal":[0],"learning":[1],"enables":[2,112],"models":[3,20,38,49,136],"to":[4,42],"process":[5],"and":[6,15,52,107,119,129,155],"reason":[7],"over":[8],"diverse":[9],"information":[10],"sources,":[11],"unlocking":[12],"human-like":[13],"perceptual":[14],"cognitive":[16],"capabilities.":[17],"As":[18],"such":[19],"gain":[21],"adoption,":[22],"efficiently":[23],"serving":[24,45,70,150],"them":[25],"on":[26,73,99,132],"GPUs":[27],"has":[28],"become":[29],"increasingly":[30],"important.":[31],"However,":[32],"the":[33,100],"modular":[34],"architecture":[35],"of":[36,127],"multimodal":[37,69,135],"poses":[39],"significant":[40],"challenges":[41],"existing":[43],"unimodal":[44],"systems,":[46],"which":[47],"treat":[48],"as":[50],"monolithic":[51],"overlook":[53],"inter-module":[54],"heterogeneity.":[55],"This":[56,110],"results":[57,143],"in":[58,152],"severe":[59],"GPU":[60,102,114,156],"underutilization.":[61],"To":[62],"address":[63],"this,":[64],"we":[65,78],"propose":[66],"Eevee,":[67],"a":[68,74,125],"system":[71],"based":[72],"new":[75],"scheduling":[76],"paradigm":[77],"call":[79],"module":[80],"multiplexing.":[81],"Unlike":[82],"prior":[83],"approaches":[84],"that":[85,145],"execute":[86],"all":[87],"modules":[88,97],"sequentially":[89],"with":[90,103],"uniform":[91],"batch":[92],"sizes,":[93],"Eevee":[94,128,146],"schedules":[95],"modality-specific":[96],"concurrently":[98],"same":[101],"independently":[104],"tuned":[105],"batching":[106],"resource":[108],"allocation.":[109],"design":[111],"fine-grained":[113],"sharing,":[115],"boosting":[116],"intra-GPU":[117],"parallelism":[118],"improving":[120],"request-level":[121],"throughput.":[122],"We":[123],"implement":[124],"prototype":[126],"evaluate":[130],"it":[131],"several":[133],"representative":[134],"(e.g.,":[137],"CLIP,":[138],"BLIP,":[139],"LLaVA,":[140],"InternVL).":[141],"Our":[142],"show":[144],"significantly":[147],"outperforms":[148],"state-of-the-art":[149],"systems":[151],"both":[153],"throughput":[154],"utilization.":[157]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2026-04-25T00:00:00"}