{"id":"https://openalex.org/W4401864094","doi":"https://doi.org/10.1145/3637528.3671465","title":"Inference Optimization of Foundation Models on AI Accelerators","display_name":"Inference Optimization of Foundation Models on AI Accelerators","publication_year":2024,"publication_date":"2024-08-24","ids":{"openalex":"https://openalex.org/W4401864094","doi":"https://doi.org/10.1145/3637528.3671465"},"language":"en","primary_location":{"id":"doi:10.1145/3637528.3671465","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3637528.3671465","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3637528.3671465","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 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3637528.3671465","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5007175636","display_name":"Youngsuk Park","orcid":"https://orcid.org/0000-0002-0970-9214"},"institutions":[],"countries":[],"is_corresponding":true,"raw_author_name":"Youngsuk Park","raw_affiliation_strings":["AWS AI, Santa Clara, CA, USA"],"raw_orcid":"https://orcid.org/0000-0002-0970-9214","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5060824941","display_name":"Kailash Budhathoki","orcid":"https://orcid.org/0000-0002-5255-8642"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Kailash Budhathoki","raw_affiliation_strings":["AWS AI, T\u00fcbingen, Germany"],"raw_orcid":"https://orcid.org/0000-0002-5255-8642","affiliations":[{"raw_affiliation_string":"AWS AI, T\u00fcbingen, Germany","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074689819","display_name":"Liangfu Chen","orcid":"https://orcid.org/0000-0002-2175-7016"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Liangfu Chen","raw_affiliation_strings":["AWS AI, Santa Clara, CA, USA"],"raw_orcid":"https://orcid.org/0000-0002-2175-7016","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5083810678","display_name":"Jonas M. K\u00fcbler","orcid":"https://orcid.org/0000-0002-8634-7990"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jonas M. K\u00fcbler","raw_affiliation_strings":["AWS AI, T\u00fcbingen, Germany"],"raw_orcid":"https://orcid.org/0000-0002-8634-7990","affiliations":[{"raw_affiliation_string":"AWS AI, T\u00fcbingen, Germany","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5082470433","display_name":"Jiaji Huang","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jiaji Huang","raw_affiliation_strings":["AWS AI, Santa Clara, CA, USA"],"raw_orcid":"https://orcid.org/0009-0006-4033-4795","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006408039","display_name":"Matth\u00e4us Kleinde\u00dfner","orcid":"https://orcid.org/0000-0002-9907-4610"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Matth\u00e4us Kleindessner","raw_affiliation_strings":["AWS AI, T\u00fcbingen, Germany"],"raw_orcid":"https://orcid.org/0000-0002-9907-4610","affiliations":[{"raw_affiliation_string":"AWS AI, T\u00fcbingen, Germany","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051326034","display_name":"Jun Huan","orcid":null},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Jun Huan","raw_affiliation_strings":["AWS AI, Santa Clara, CA, USA"],"raw_orcid":"https://orcid.org/0009-0007-6440-4732","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, CA, USA","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5027059837","display_name":"Volkan Cevher","orcid":"https://orcid.org/0000-0002-5004-201X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Volkan Cevher","raw_affiliation_strings":["AWS AI & EPFL, T\u00fcbingen, Germany"],"raw_orcid":"https://orcid.org/0000-0002-5004-201X","affiliations":[{"raw_affiliation_string":"AWS AI & EPFL, T\u00fcbingen, Germany","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101638214","display_name":"Yida Wang","orcid":"https://orcid.org/0000-0001-8165-840X"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Yida Wang","raw_affiliation_strings":["AWS AI, Santa Clara, CA, USA"],"raw_orcid":"https://orcid.org/0000-0001-8165-840X","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, CA, USA","institution_ids":[]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5082384108","display_name":"George Karypis","orcid":"https://orcid.org/0000-0003-2753-1437"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"George Karypis","raw_affiliation_strings":["AWS AI, Santa Clara, CA, USA"],"raw_orcid":"https://orcid.org/0000-0003-2753-1437","affiliations":[{"raw_affiliation_string":"AWS AI, Santa Clara, CA, USA","institution_ids":[]}]}],"institutions":[],"countries_distinct_count":0,"institutions_distinct_count":10,"corresponding_author_ids":["https://openalex.org/A5007175636"],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.9275,"has_fulltext":false,"cited_by_count":6,"citation_normalized_percentile":{"value":0.84464469,"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":"6605","last_page":"6615"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11801","display_name":"Reservoir Engineering and Simulation Methods","score":0.9746000170707703,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11801","display_name":"Reservoir Engineering and Simulation Methods","score":0.9746000170707703,"subfield":{"id":"https://openalex.org/subfields/2212","display_name":"Ocean Engineering"},"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/T10320","display_name":"Neural Networks and Applications","score":0.942300021648407,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.927299976348877,"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/inference","display_name":"Inference","score":0.6885156631469727},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6590360403060913},{"id":"https://openalex.org/keywords/foundation","display_name":"Foundation (evidence)","score":0.5585445165634155},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4446585178375244},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.346706748008728}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.6885156631469727},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6590360403060913},{"id":"https://openalex.org/C2780966255","wikidata":"https://www.wikidata.org/wiki/Q5474306","display_name":"Foundation (evidence)","level":2,"score":0.5585445165634155},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4446585178375244},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.346706748008728},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3637528.3671465","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3637528.3671465","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3637528.3671465","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 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3637528.3671465","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3637528.3671465","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3637528.3671465","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 30th ACM SIGKDD Conference on Knowledge Discovery and Data Mining","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":true,"grobid_xml":false},"content_urls":{"pdf":"https://content.openalex.org/works/W4401864094.pdf"},"referenced_works_count":26,"referenced_works":["https://openalex.org/W2294370754","https://openalex.org/W2606722458","https://openalex.org/W2612387305","https://openalex.org/W2888727064","https://openalex.org/W2963122961","https://openalex.org/W3047848469","https://openalex.org/W3119866685","https://openalex.org/W3137147200","https://openalex.org/W4210247494","https://openalex.org/W4212774754","https://openalex.org/W4226079124","https://openalex.org/W4226126941","https://openalex.org/W4237729708","https://openalex.org/W4281758439","https://openalex.org/W4287391717","https://openalex.org/W4312933868","https://openalex.org/W4321636575","https://openalex.org/W4380874786","https://openalex.org/W4385571586","https://openalex.org/W4385573023","https://openalex.org/W4386065704","https://openalex.org/W4386071831","https://openalex.org/W4387321091","https://openalex.org/W4395110738","https://openalex.org/W4402672007","https://openalex.org/W6810737565"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W3046775127","https://openalex.org/W3107602296","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"Powerful":[0],"foundation":[1,38],"models,":[2],"including":[3],"large":[4,30],"language":[5],"models":[6],"(LLMs),":[7],"with":[8,116],"Transformer":[9,121],"architectures":[10,122],"have":[11,27],"ushered":[12],"in":[13,78,174],"a":[14,29,82,104],"new":[15,33],"era":[16],"of":[17,32,61,67,119],"Generative":[18],"AI":[19,92,113,150],"across":[20],"various":[21,167],"industries.":[22],"Industry":[23],"and":[24,44,49,52,75,88,123,137,141,170],"research":[25],"community":[26],"witnessed":[28],"number":[31,60],"applications,":[34],"based":[35],"on":[36,107,149],"those":[37],"models.":[39],"Such":[40],"applications":[41],"include":[42],"question":[43],"answer,":[45],"customer":[46],"services,":[47],"image":[48],"video":[50],"generation,":[51],"code":[53],"completions,":[54],"among":[55],"others.":[56],"However,":[57],"as":[58],"the":[59,84,175],"model":[62,168],"parameters":[63],"reaches":[64],"to":[65],"hundreds":[66],"billions,":[68],"their":[69],"deployment":[70],"incurs":[71],"prohibitive":[72],"inference":[73,90,109],"costs":[74],"high":[76],"latency":[77],"real-world":[79],"scenarios.":[80],"As":[81],"result,":[83],"demand":[85],"for":[86,135,160],"cost-effective":[87],"fast":[89,136,161,171],"using":[91,112],"accelerators":[93],"is":[94],"ever":[95],"more":[96],"higher.":[97],"To":[98],"this":[99],"end,":[100],"our":[101],"tutorial":[102],"offers":[103],"comprehensive":[105],"discussion":[106],"complementary":[108],"optimization":[110,133],"techniques":[111,134],"accelerators.":[114,151],"Beginning":[115],"an":[117],"overview":[118],"basic":[120],"deep":[124,129],"learning":[125],"system":[126,132],"frameworks,":[127],"we":[128,153,165],"dive":[130],"into":[131],"memory-efficient":[138],"attention":[139],"computations":[140],"discuss":[142],"how":[143],"they":[144],"can":[145],"be":[146],"implemented":[147],"efficiently":[148],"Next,":[152],"describe":[154],"architectural":[155],"elements":[156],"that":[157],"are":[158],"key":[159],"transformer":[162],"inference.":[163],"Finally,":[164],"examine":[166],"compression":[169],"decoding":[172],"strategies":[173],"same":[176],"context.":[177]},"counts_by_year":[{"year":2025,"cited_by_count":5},{"year":2024,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}