{"id":"https://openalex.org/W4417243463","doi":"https://doi.org/10.1145/3760250.3762220","title":"MoDM: Efficient Serving for Image Generation via <u>M</u> ixture- <u>o</u> f- <u>D</u> iffusion <u>M</u> odels","display_name":"MoDM: Efficient Serving for Image Generation via <u>M</u> ixture- <u>o</u> f- <u>D</u> iffusion <u>M</u> odels","publication_year":2025,"publication_date":"2025-12-11","ids":{"openalex":"https://openalex.org/W4417243463","doi":"https://doi.org/10.1145/3760250.3762220"},"language":null,"primary_location":{"id":"doi:10.1145/3760250.3762220","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3760250.3762220","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3760250.3762220","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3760250.3762220","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5111322449","display_name":"Yuchen Xia","orcid":"https://orcid.org/0009-0002-5557-3095"},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yuchen Xia","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":"https://orcid.org/0009-0002-5557-3095","affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5030023429","display_name":"Divyam Sharma","orcid":"https://orcid.org/0009-0008-3740-7353"},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Divyam Sharma","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":"https://orcid.org/0009-0008-3740-7353","affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yichao Yuan","orcid":"https://orcid.org/0000-0003-0509-8687"},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Yichao Yuan","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":"https://orcid.org/0000-0003-0509-8687","affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5076002956","display_name":"Souvik Kundu","orcid":"https://orcid.org/0000-0002-3533-9405"},"institutions":[{"id":"https://openalex.org/I1343180700","display_name":"Intel (United States)","ror":"https://ror.org/01ek73717","country_code":"US","type":"company","lineage":["https://openalex.org/I1343180700"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Souvik Kundu","raw_affiliation_strings":["Intel Labs, Los Angeles, CA, USA"],"raw_orcid":"https://orcid.org/0000-0002-3533-9405","affiliations":[{"raw_affiliation_string":"Intel Labs, Los Angeles, CA, USA","institution_ids":["https://openalex.org/I1343180700"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5063661349","display_name":"Nishil Talati","orcid":"https://orcid.org/0000-0002-2457-4119"},"institutions":[{"id":"https://openalex.org/I27837315","display_name":"University of Michigan","ror":"https://ror.org/00jmfr291","country_code":"US","type":"education","lineage":["https://openalex.org/I27837315"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Nishil Talati","raw_affiliation_strings":["University of Michigan, Ann Arbor, MI, USA"],"raw_orcid":"https://orcid.org/0000-0002-2457-4119","affiliations":[{"raw_affiliation_string":"University of Michigan, Ann Arbor, MI, USA","institution_ids":["https://openalex.org/I27837315"]}]}],"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":true,"cited_by_count":0,"citation_normalized_percentile":null,"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"163","last_page":"182"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.367900013923645,"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/T10775","display_name":"Generative Adversarial Networks and Image Synthesis","score":0.367900013923645,"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/T10481","display_name":"Computer Graphics and Visualization Techniques","score":0.16920000314712524,"subfield":{"id":"https://openalex.org/subfields/1704","display_name":"Computer Graphics and Computer-Aided Design"},"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.07919999957084656,"subfield":{"id":"https://openalex.org/subfields/1705","display_name":"Computer Networks and Communications"},"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/scalability","display_name":"Scalability","score":0.7490000128746033},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.670199990272522},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5968999862670898},{"id":"https://openalex.org/keywords/reuse","display_name":"Reuse","score":0.5461999773979187},{"id":"https://openalex.org/keywords/image-quality","display_name":"Image quality","score":0.4916999936103821},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.4912000000476837},{"id":"https://openalex.org/keywords/image","display_name":"Image (mathematics)","score":0.47099998593330383},{"id":"https://openalex.org/keywords/throughput","display_name":"Throughput","score":0.39629998803138733}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8274999856948853},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7490000128746033},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.670199990272522},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5968999862670898},{"id":"https://openalex.org/C206588197","wikidata":"https://www.wikidata.org/wiki/Q846574","display_name":"Reuse","level":2,"score":0.5461999773979187},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.49729999899864197},{"id":"https://openalex.org/C55020928","wikidata":"https://www.wikidata.org/wiki/Q3813865","display_name":"Image quality","level":3,"score":0.4916999936103821},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.4912000000476837},{"id":"https://openalex.org/C115961682","wikidata":"https://www.wikidata.org/wiki/Q860623","display_name":"Image (mathematics)","level":2,"score":0.47099998593330383},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.4235000014305115},{"id":"https://openalex.org/C157764524","wikidata":"https://www.wikidata.org/wiki/Q1383412","display_name":"Throughput","level":3,"score":0.39629998803138733},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3515999913215637},{"id":"https://openalex.org/C1667742","wikidata":"https://www.wikidata.org/wiki/Q10927554","display_name":"Image retrieval","level":3,"score":0.33719998598098755},{"id":"https://openalex.org/C5119721","wikidata":"https://www.wikidata.org/wiki/Q220501","display_name":"Quality of service","level":2,"score":0.33169999718666077},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.3310999870300293},{"id":"https://openalex.org/C9417928","wikidata":"https://www.wikidata.org/wiki/Q1070689","display_name":"Image processing","level":3,"score":0.3156000077724457},{"id":"https://openalex.org/C177774035","wikidata":"https://www.wikidata.org/wiki/Q1246948","display_name":"Granularity","level":2,"score":0.3125999867916107},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3012000024318695},{"id":"https://openalex.org/C31972630","wikidata":"https://www.wikidata.org/wiki/Q844240","display_name":"Computer vision","level":1,"score":0.27799999713897705},{"id":"https://openalex.org/C189783530","wikidata":"https://www.wikidata.org/wiki/Q352090","display_name":"CPU cache","level":3,"score":0.25450000166893005},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.2533000111579895},{"id":"https://openalex.org/C124101348","wikidata":"https://www.wikidata.org/wiki/Q172491","display_name":"Data mining","level":1,"score":0.25279998779296875},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.2515999972820282},{"id":"https://openalex.org/C46637626","wikidata":"https://www.wikidata.org/wiki/Q6693015","display_name":"Low latency (capital markets)","level":2,"score":0.25}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3760250.3762220","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3760250.3762220","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3760250.3762220","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3760250.3762220","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3760250.3762220","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3760250.3762220","source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 31st ACM International Conference on Architectural Support for Programming Languages and Operating Systems, Volume 1","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4417243463.pdf","grobid_xml":"https://content.openalex.org/works/W4417243463.grobid-xml"},"referenced_works_count":5,"referenced_works":["https://openalex.org/W2183341477","https://openalex.org/W3153469116","https://openalex.org/W4385570204","https://openalex.org/W4394998532","https://openalex.org/W4402816687"],"related_works":[],"abstract_inverted_index":{"Diffusion-based":[0],"text-to-image":[1],"generation":[2],"models":[3,9,19,36,86,96],"trade":[4],"latency":[5,40,80,92],"for":[6,34,87,97,160],"quality:":[7,83],"small":[8],"are":[10,24],"fast":[11],"but":[12,23],"generate":[13],"lower":[14],"quality":[15,42,106],"images,":[16,61],"while":[17,93,130,151],"large":[18],"produce":[20],"better":[21],"images":[22],"slow.":[25],"We":[26,113],"present":[27],"MoDM,":[28],"a":[29,44,115,157],"novel":[30],"caching-based":[31],"serving":[32,76,147],"system":[33],"diffusion":[35,47,69],"that":[37,52,118,142],"dynamically":[38,78],"balances":[39,124],"and":[41,65,81,123,162],"through":[43],"mixture":[45],"of":[46],"models.":[48],"Unlike":[49],"prior":[50],"approaches":[51],"rely":[53],"on":[54],"model-specific":[55],"internal":[56],"features,":[57],"MoDM":[58,143],"caches":[59],"final":[60],"allowing":[62],"seamless":[63],"retrieval":[64],"reuse":[66],"across":[67],"multiple":[68],"model":[70,104,164],"families.":[71],"This":[72],"design":[73,114],"enables":[74],"adaptive":[75],"by":[77,149],"balancing":[79],"image":[82,105,153],"using":[84,109],"smaller":[85],"cache-hit":[88],"requests":[89,99],"to":[90,100],"reduce":[91],"reserving":[94],"larger":[95],"cache-miss":[98],"maintain":[101],"quality.":[102],"Small":[103],"is":[107],"preserved":[108],"retrieved":[110],"cached":[111],"images.":[112],"global":[116],"monitor":[117],"optimally":[119],"allocates":[120],"GPU":[121],"resources":[122],"inference":[125],"workload,":[126],"ensuring":[127],"high":[128],"throughput":[129],"meeting":[131],"Service-Level":[132],"Objectives":[133],"(SLOs)":[134],"under":[135],"varying":[136],"request":[137],"rates.":[138],"Our":[139],"evaluations":[140],"show":[141],"significantly":[144],"reduces":[145],"average":[146],"time":[148],"2.5\u00d7":[150],"retaining":[152],"quality,":[154],"making":[155],"it":[156],"practical":[158],"solution":[159],"scalable":[161],"resource-efficient":[163],"deployment.":[165]},"counts_by_year":[],"updated_date":"2026-07-14T23:27:15.235271","created_date":"2025-12-11T00:00:00"}