{"id":"https://openalex.org/W4415708213","doi":"https://doi.org/10.1109/icme59968.2025.11209145","title":"InterLayer: Efficient Inference with Interleaved Scheduling and Layer-Specific Optimization","display_name":"InterLayer: Efficient Inference with Interleaved Scheduling and Layer-Specific Optimization","publication_year":2025,"publication_date":"2025-06-30","ids":{"openalex":"https://openalex.org/W4415708213","doi":"https://doi.org/10.1109/icme59968.2025.11209145"},"language":null,"primary_location":{"id":"doi:10.1109/icme59968.2025.11209145","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icme59968.2025.11209145","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","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/A5100354225","display_name":"Cheng Li","orcid":"https://orcid.org/0000-0001-6110-8099"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Limin Cheng","raw_affiliation_strings":["Chinese Academy of Sciences,Institute of Software"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,Institute of Software","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5003549015","display_name":"Hang Qin","orcid":"https://orcid.org/0000-0002-7435-9652"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hang Qin","raw_affiliation_strings":["Chinese Academy of Sciences,Institute of Software"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,Institute of Software","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Shouxu Kuang","orcid":null},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Shouxu Kuang","raw_affiliation_strings":["Chinese Academy of Sciences,Institute of Software"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,Institute of Software","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100352839","display_name":"Xinyu Wang","orcid":"https://orcid.org/0000-0003-4143-334X"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xinyu Wang","raw_affiliation_strings":["Chinese Academy of Sciences,Institute of Software"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,Institute of Software","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5037788925","display_name":"Ling Li","orcid":"https://orcid.org/0000-0001-9722-9503"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ling Li","raw_affiliation_strings":["Chinese Academy of Sciences,Institute of Software"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,Institute of Software","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101053493","display_name":"Yanjun Wu","orcid":"https://orcid.org/0000-0002-1823-0459"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yanjun Wu","raw_affiliation_strings":["Chinese Academy of Sciences,Institute of Software"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,Institute of Software","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5101669954","display_name":"Chen Zhao","orcid":"https://orcid.org/0000-0003-4226-7793"},"institutions":[{"id":"https://openalex.org/I19820366","display_name":"Chinese Academy of Sciences","ror":"https://ror.org/034t30j35","country_code":"CN","type":"government","lineage":["https://openalex.org/I19820366"]},{"id":"https://openalex.org/I4210128818","display_name":"Institute of Software","ror":"https://ror.org/033dfsn42","country_code":"CN","type":"facility","lineage":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Chen Zhao","raw_affiliation_strings":["Chinese Academy of Sciences,Institute of Software"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"Chinese Academy of Sciences,Institute of Software","institution_ids":["https://openalex.org/I19820366","https://openalex.org/I4210128818"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":2,"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":"1","last_page":"6"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T14347","display_name":"Big Data and Digital Economy","score":0.19339999556541443,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T14347","display_name":"Big Data and Digital Economy","score":0.19339999556541443,"subfield":{"id":"https://openalex.org/subfields/1710","display_name":"Information Systems"},"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/T10036","display_name":"Advanced Neural Network Applications","score":0.12160000205039978,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.05590000003576279,"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/inference","display_name":"Inference","score":0.7444999814033508},{"id":"https://openalex.org/keywords/scheduling","display_name":"Scheduling (production processes)","score":0.72079998254776},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.6632999777793884},{"id":"https://openalex.org/keywords/computation","display_name":"Computation","score":0.6484000086784363},{"id":"https://openalex.org/keywords/server","display_name":"Server","score":0.6097999811172485},{"id":"https://openalex.org/keywords/cloud-computing","display_name":"Cloud computing","score":0.5828999876976013},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.33399999141693115},{"id":"https://openalex.org/keywords/cuda","display_name":"CUDA","score":0.3255000114440918}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8378999829292297},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7444999814033508},{"id":"https://openalex.org/C206729178","wikidata":"https://www.wikidata.org/wiki/Q2271896","display_name":"Scheduling (production processes)","level":2,"score":0.72079998254776},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.6632999777793884},{"id":"https://openalex.org/C45374587","wikidata":"https://www.wikidata.org/wiki/Q12525525","display_name":"Computation","level":2,"score":0.6484000086784363},{"id":"https://openalex.org/C93996380","wikidata":"https://www.wikidata.org/wiki/Q44127","display_name":"Server","level":2,"score":0.6097999811172485},{"id":"https://openalex.org/C79974875","wikidata":"https://www.wikidata.org/wiki/Q483639","display_name":"Cloud computing","level":2,"score":0.5828999876976013},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4544000029563904},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.36250001192092896},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.33399999141693115},{"id":"https://openalex.org/C2778119891","wikidata":"https://www.wikidata.org/wiki/Q477690","display_name":"CUDA","level":2,"score":0.3255000114440918},{"id":"https://openalex.org/C2781041963","wikidata":"https://www.wikidata.org/wiki/Q18348618","display_name":"Computation offloading","level":4,"score":0.3158000111579895},{"id":"https://openalex.org/C39890363","wikidata":"https://www.wikidata.org/wiki/Q36108","display_name":"Generative grammar","level":2,"score":0.30809998512268066},{"id":"https://openalex.org/C137836250","wikidata":"https://www.wikidata.org/wiki/Q984063","display_name":"Optimization problem","level":2,"score":0.3068999946117401},{"id":"https://openalex.org/C167966045","wikidata":"https://www.wikidata.org/wiki/Q5532625","display_name":"Generative model","level":3,"score":0.30550000071525574},{"id":"https://openalex.org/C50630238","wikidata":"https://www.wikidata.org/wiki/Q971505","display_name":"General-purpose computing on graphics processing units","level":3,"score":0.2892000079154968},{"id":"https://openalex.org/C83283714","wikidata":"https://www.wikidata.org/wiki/Q121117","display_name":"Supercomputer","level":2,"score":0.2854999899864197},{"id":"https://openalex.org/C2984822820","wikidata":"https://www.wikidata.org/wiki/Q1123036","display_name":"Processor scheduling","level":3,"score":0.26249998807907104},{"id":"https://openalex.org/C116537","wikidata":"https://www.wikidata.org/wiki/Q2169973","display_name":"Service provider","level":3,"score":0.2599000036716461},{"id":"https://openalex.org/C175154964","wikidata":"https://www.wikidata.org/wiki/Q380077","display_name":"Task analysis","level":3,"score":0.25290000438690186}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/icme59968.2025.11209145","is_oa":false,"landing_page_url":"https://doi.org/10.1109/icme59968.2025.11209145","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 IEEE International Conference on Multimedia and Expo (ICME)","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":12,"referenced_works":["https://openalex.org/W2944614352","https://openalex.org/W4372263604","https://openalex.org/W4380925941","https://openalex.org/W4386160289","https://openalex.org/W4387321091","https://openalex.org/W4390263770","https://openalex.org/W4401408839","https://openalex.org/W4401991019","https://openalex.org/W4402980933","https://openalex.org/W4404356490","https://openalex.org/W4404385775","https://openalex.org/W4404401018"],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"and":[4,66,69,89,123,129,140],"multimodal":[5],"LLMs":[6,31,122],"(MLLMs)":[7],"are":[8],"extensively":[9],"used":[10],"in":[11],"latency-sensitive":[12],"multimedia":[13],"applications,":[14],"with":[15],"inference":[16,58],"services":[17],"typically":[18],"offered":[19],"by":[20],"generative":[21],"AI":[22],"(GenAI)":[23],"cloud":[24],"service":[25],"providers.":[26],"However,":[27],"switching":[28],"between":[29],"different":[30],"or":[32],"MLLMs":[33,124],"on":[34],"GPU":[35,42,75,94,104],"servers":[36],"for":[37,120],"diverse":[38,126],"user":[39],"needs":[40],"causes":[41,73],"computation":[43,76,105],"stalls,":[44],"leading":[45],"to":[46,92,133,142],"high":[47],"cold-start":[48],"latency.":[49],"To":[50],"address":[51],"these":[52],"challenges,":[53],"we":[54,61,79,97],"propose":[55],"an":[56],"efficient":[57],"scheme.":[59],"First,":[60],"analyze":[62,99],"the":[63,71,100,146],"layer-wise":[64],"loading":[65],"execution":[67],"characteristics":[68],"identify":[70],"root":[72],"of":[74,103],"stalls.":[77,95],"Then,":[78],"introduce":[80],"InterLayer,":[81],"leveraging":[82],"fine-grained,":[83],"compute-dependency-aware":[84],"interleaved":[85],"scheduling":[86],"parallel":[87],"transmission":[88],"layer-specific":[90],"optimization":[91],"reduce":[93],"Additionally,":[96],"theoretically":[98],"lower":[101],"bound":[102],"stalls":[106],"using":[107],"InterLayer.":[108],"Experimental":[109],"results":[110],"show":[111],"that":[112],"InterLayer":[113],"achieves":[114],"a":[115,134],"2.4\u20134.9\u00d7":[116],"acceleration":[117],"over":[118,138,145],"vLLM":[119],"multiple":[121],"across":[125],"model":[127],"architectures":[128],"hardware":[130],"configurations,":[131],"up":[132,141],"3.6\u00d7":[135],"throughput":[136],"improvement":[137],"vLLM,":[139],"2.2\u00d7":[143],"speedup":[144],"latest":[147],"state-of-the-art.":[148]},"counts_by_year":[],"updated_date":"2026-07-22T07:51:19.307946","created_date":"2025-10-30T00:00:00"}