{"id":"https://openalex.org/W6903240423","doi":"https://doi.org/10.1145/3725843.3756121","title":"Pimba: A Processing-in-Memory Acceleration for Post-Transformer Large Language Model Serving","display_name":"Pimba: A Processing-in-Memory Acceleration for Post-Transformer Large Language Model Serving","publication_year":2025,"publication_date":"2025-10-17","ids":{"openalex":"https://openalex.org/W6903240423","doi":"https://doi.org/10.1145/3725843.3756121"},"language":"en","primary_location":{"id":"doi:10.1145/3725843.3756121","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725843.3756121","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3725843.3756121","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 58th IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},"type":"conference-paper","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3725843.3756121","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Wonung Kim","orcid":"https://orcid.org/0009-0000-1180-0503"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Wonung Kim","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0000-1180-0503","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yubin Lee","orcid":"https://orcid.org/0009-0002-6541-6739"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yubin Lee","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0002-6541-6739","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yoonsung Kim","orcid":"https://orcid.org/0009-0000-2333-292X"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yoonsung Kim","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0000-2333-292X","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jinwoo Hwang","orcid":"https://orcid.org/0009-0008-8498-2502"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jinwoo Hwang","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0008-8498-2502","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Seongryong Oh","orcid":"https://orcid.org/0009-0004-6707-0641"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seongryong Oh","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0004-6707-0641","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jiyong Jung","orcid":"https://orcid.org/0009-0007-0420-179X"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jiyong Jung","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0007-0420-179X","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Aziz Huseynov","orcid":"https://orcid.org/0009-0009-0516-0605"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Aziz Huseynov","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0009-0516-0605","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Woong Gyu Park","orcid":"https://orcid.org/0009-0002-4106-8039"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Woong Gyu Park","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0002-4106-8039","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Chang Hyun Park","orcid":"https://orcid.org/0000-0002-8250-8574"},"institutions":[{"id":"https://openalex.org/I123387679","display_name":"Uppsala University","ror":"https://ror.org/048a87296","country_code":"SE","type":"education","lineage":["https://openalex.org/I123387679"]}],"countries":["SE"],"is_corresponding":false,"raw_author_name":"Chang Hyun Park","raw_affiliation_strings":["Uppsala University, Atlanta, GA, USA"],"raw_orcid":"https://orcid.org/0000-0002-8250-8574","affiliations":[{"raw_affiliation_string":"Uppsala University, Atlanta, GA, USA","institution_ids":["https://openalex.org/I123387679"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Divya Mahajan","orcid":"https://orcid.org/0009-0007-8184-0528"},"institutions":[{"id":"https://openalex.org/I130701444","display_name":"Georgia Institute of Technology","ror":"https://ror.org/01zkghx44","country_code":"US","type":"education","lineage":["https://openalex.org/I130701444"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Divya Mahajan","raw_affiliation_strings":["Georgia Institute of Technology, Uppsala, Sweden"],"raw_orcid":"https://orcid.org/0009-0007-8184-0528","affiliations":[{"raw_affiliation_string":"Georgia Institute of Technology, Uppsala, Sweden","institution_ids":["https://openalex.org/I130701444"]}]},{"author_position":"last","author":{"id":null,"display_name":"Jongse Park","orcid":"https://orcid.org/0000-0002-6629-449X"},"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"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jongse Park","raw_affiliation_strings":["KAIST, Daejeon, Republic of Korea"],"raw_orcid":"https://orcid.org/0000-0002-6629-449X","affiliations":[{"raw_affiliation_string":"KAIST, Daejeon, Republic of Korea","institution_ids":["https://openalex.org/I157485424"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6199,"has_fulltext":true,"cited_by_count":2,"citation_normalized_percentile":{"value":0.87877703,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"292","last_page":"307"},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.2502000033855438,"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"}},"topics":[{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.2502000033855438,"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/T10028","display_name":"Topic Modeling","score":0.2142000049352646,"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/T14347","display_name":"Big Data and Digital Economy","score":0.06040000170469284,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.7264000177383423},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.6147000193595886},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5044999718666077},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.49390000104904175},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.42410001158714294},{"id":"https://openalex.org/keywords/execution-model","display_name":"Execution model","score":0.3984000086784363},{"id":"https://openalex.org/keywords/adder","display_name":"Adder","score":0.39430001378059387},{"id":"https://openalex.org/keywords/multiplier","display_name":"Multiplier (economics)","score":0.35429999232292175},{"id":"https://openalex.org/keywords/inference-engine","display_name":"Inference engine","score":0.35370001196861267}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7594000101089478},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.7264000177383423},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.6147000193595886},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5044999718666077},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.49390000104904175},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.42410001158714294},{"id":"https://openalex.org/C2776834041","wikidata":"https://www.wikidata.org/wiki/Q25346349","display_name":"Execution model","level":2,"score":0.3984000086784363},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3959999978542328},{"id":"https://openalex.org/C164620267","wikidata":"https://www.wikidata.org/wiki/Q376953","display_name":"Adder","level":3,"score":0.39430001378059387},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.38749998807907104},{"id":"https://openalex.org/C124584101","wikidata":"https://www.wikidata.org/wiki/Q1053266","display_name":"Multiplier (economics)","level":2,"score":0.35429999232292175},{"id":"https://openalex.org/C46743427","wikidata":"https://www.wikidata.org/wiki/Q1341685","display_name":"Inference engine","level":3,"score":0.35370001196861267},{"id":"https://openalex.org/C48103436","wikidata":"https://www.wikidata.org/wiki/Q599031","display_name":"State (computer science)","level":2,"score":0.34769999980926514},{"id":"https://openalex.org/C120314980","wikidata":"https://www.wikidata.org/wiki/Q180634","display_name":"Distributed computing","level":1,"score":0.34619998931884766},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.3100999891757965},{"id":"https://openalex.org/C44291984","wikidata":"https://www.wikidata.org/wiki/Q1074173","display_name":"Question answering","level":2,"score":0.30979999899864197},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.3003000020980835},{"id":"https://openalex.org/C167822520","wikidata":"https://www.wikidata.org/wiki/Q176452","display_name":"Finite-state machine","level":2,"score":0.28610000014305115},{"id":"https://openalex.org/C12186640","wikidata":"https://www.wikidata.org/wiki/Q6815743","display_name":"Memory model","level":3,"score":0.28540000319480896},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.28529998660087585},{"id":"https://openalex.org/C2780069185","wikidata":"https://www.wikidata.org/wiki/Q7977945","display_name":"Equivalence (formal languages)","level":2,"score":0.27900001406669617},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.2768000066280365},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2759999930858612},{"id":"https://openalex.org/C2777767291","wikidata":"https://www.wikidata.org/wiki/Q1080291","display_name":"Sizing","level":2,"score":0.2644999921321869},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.2639999985694885},{"id":"https://openalex.org/C19275194","wikidata":"https://www.wikidata.org/wiki/Q222903","display_name":"Multiplexing","level":2,"score":0.262800008058548},{"id":"https://openalex.org/C72434380","wikidata":"https://www.wikidata.org/wiki/Q230930","display_name":"State space","level":2,"score":0.259799987077713}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3725843.3756121","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725843.3756121","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3725843.3756121","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 58th IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},{"id":"pmh:oai:arXiv.org:2507.10178","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2507.10178","pdf_url":"https://arxiv.org/pdf/2507.10178","source":{"id":"https://openalex.org/S4306400194","display_name":"arXiv (Cornell University)","issn_l":"2331-8422","issn":["2331-8422"],"is_oa":true,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I205783295","host_organization_name":"Cornell University","host_organization_lineage":["https://openalex.org/I205783295"],"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":"text"},{"id":"pmh:oai:DiVA.org:uu-574839","is_oa":true,"landing_page_url":"http://urn.kb.se/resolve?urn=urn:nbn:se:uu:diva-574839","pdf_url":"https://uu.diva-portal.org/smash/get/diva2:2026046/FULLTEXT01","source":{"id":"https://openalex.org/S4306400653","display_name":"Diva portal (Dalarna University Library)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":null,"host_organization_name":null,"host_organization_lineage":[],"host_organization_lineage_names":[],"type":"repository"},"license":"other-oa","license_id":"https://openalex.org/licenses/other-oa","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"info:eu-repo/semantics/conferenceObject"}],"best_oa_location":{"id":"doi:10.1145/3725843.3756121","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725843.3756121","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3725843.3756121","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 58th IEEE/ACM International Symposium on Microarchitecture","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G2638105757","display_name":null,"funder_award_id":"RS-2025-02305453","funder_id":"https://openalex.org/F4320322093","funder_display_name":"Electronics and Telecommunications Research Institute"},{"id":"https://openalex.org/G3812690899","display_name":null,"funder_award_id":"RS-2023","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G4603601854","display_name":null,"funder_award_id":"RS-2025-02305453","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G4700831490","display_name":null,"funder_award_id":"2022-","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G5906391025","display_name":null,"funder_award_id":"2022-0-01037","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"}],"funders":[{"id":"https://openalex.org/F4320307601","display_name":"Idaho Humanities Council","ror":"https://ror.org/02jjxqt79"},{"id":"https://openalex.org/F4320322093","display_name":"Electronics and Telecommunications Research Institute","ror":"https://ror.org/03ysstz10"},{"id":"https://openalex.org/F4320322202","display_name":"IC Design Education Center","ror":"https://ror.org/005v57z85"},{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"},{"id":"https://openalex.org/F4320335489","display_name":"Institute for Information and Communications Technology Promotion","ror":"https://ror.org/01g0hqq23"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W6903240423.pdf","grobid_xml":"https://content.openalex.org/works/W6903240423.grobid-xml"},"referenced_works_count":24,"referenced_works":["https://openalex.org/W2053987390","https://openalex.org/W2120000030","https://openalex.org/W2725159389","https://openalex.org/W2761132374","https://openalex.org/W2946609015","https://openalex.org/W2963015836","https://openalex.org/W2998617917","https://openalex.org/W3100710793","https://openalex.org/W3133253223","https://openalex.org/W3158634533","https://openalex.org/W3194676777","https://openalex.org/W3208633927","https://openalex.org/W4387321091","https://openalex.org/W4388757726","https://openalex.org/W4389524555","https://openalex.org/W4392427708","https://openalex.org/W4393407040","https://openalex.org/W4393407316","https://openalex.org/W4394998532","https://openalex.org/W4394998968","https://openalex.org/W4395073431","https://openalex.org/W4404848672","https://openalex.org/W4404955380","https://openalex.org/W4412888538"],"related_works":[],"abstract_inverted_index":{"Transformers":[0],"are":[1,105],"the":[2,13,36,93,136,194],"driving":[3],"force":[4],"behind":[5],"today's":[6],"Large":[7],"Language":[8],"Models":[9],"(LLMs),":[10],"serving":[11,69],"as":[12,43,60,185],"foundation":[14],"for":[15,32],"their":[16,20,101],"performance":[17,94],"and":[18,22,52,98,118,146,170],"versatility.Yet,":[19],"compute":[21],"memory":[23,109,137],"costs":[24],"grow":[25],"with":[26],"sequence":[27],"length,":[28],"posing":[29],"scalability":[30],"challenges":[31],"long-context":[33],"inferencing.In":[34],"response,":[35],"algorithm":[37],"community":[38],"is":[39],"exploring":[40],"alternative":[41],"architectures-such":[42],"state":[44,119,156],"space":[45],"models":[46,83],"(SSMs)":[47],"(e.g.,":[48],"Mamba-2),":[49],"linear":[50],"attention,":[51,160],"recurrent":[53],"neural":[54],"networks":[55],"(RNNs)-which":[56],"we":[57,91,126,181,192],"refer":[58],"to":[59,114,134],"post-transformers.This":[61],"shift":[62],"presents":[63],"a":[64,68,85,186],"key":[65],"challenge:":[66],"building":[67],"system":[70],"that":[71,132],"efficiently":[72],"supports":[73],"not":[74],"only":[75],"emerging":[76],"post-transformer":[77,99],"LLMs":[78],"but":[79],"also":[80],"existing":[81],"transformer":[82,97],"within":[84],"unified":[86],"framework.To":[87],"address":[88,135],"this":[89,124],"challenge,":[90],"analyze":[92],"characteristics":[95],"of":[96],"LLMs.Despite":[100],"algorithmic":[102],"differences,":[103],"both":[104],"largely":[106],"bounded":[107],"by":[108,123,139],"bandwidth":[110],"under":[111],"batched":[112],"inference-due":[113],"attention":[115],"in":[116,121],"transformers":[117],"updates":[120],"post-transformers.Inspired":[122],"finding,":[125],"propose":[127],"Pimba,":[128],"an":[129],"accelerator":[130],"solution":[131],"aims":[133],"bottleneck":[138],"jointly":[140],"leveraging":[141],"(1)":[142,155],"Processing-in-Memory":[143],"(PIM)":[144],"paradigm":[145],"(2)":[147,171],"LLM":[148],"quantization.Further":[149],"analyses":[150],"suggest":[151],"two":[152],"additional":[153],"insights:":[154],"update":[157],"operations,":[158],"unlike":[159],"incur":[161],"high":[162],"hardware":[163],"cost,":[164],"making":[165],"per-bank":[166],"PIM":[167],"acceleration":[168],"inefficient,":[169],"different":[172],"low-precision":[173],"arithmetic":[174],"methods":[175],"offer":[176],"varying":[177],"accuracy-area":[178],"tradeoffs,":[179],"while":[180],"identify":[182],"Microsoft's":[183],"MX":[184],"Pareto-optimal":[187],"choice.Building":[188],"on":[189],"these":[190],"insights,":[191],"design":[193]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-08-08T01:25:22.217667","created_date":"2025-10-10T00:00:00"}