{"id":"https://openalex.org/W7119941304","doi":"https://doi.org/10.1109/mc.2026.3652916","title":"Challenges and Research Directions for Large Language Model Inference Hardware","display_name":"Challenges and Research Directions for Large Language Model Inference Hardware","publication_year":2026,"publication_date":"2026-05-01","ids":{"openalex":"https://openalex.org/W7119941304","doi":"https://doi.org/10.1109/mc.2026.3652916"},"language":null,"primary_location":{"id":"doi:10.1109/mc.2026.3652916","is_oa":true,"landing_page_url":"https://doi.org/10.1109/mc.2026.3652916","pdf_url":null,"source":{"id":"https://openalex.org/S178916657","display_name":"Computer","issn_l":"0018-9162","issn":["0018-9162","1558-0814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer","raw_type":"journal-article"},"type":"article","indexed_in":["arxiv","crossref"],"open_access":{"is_oa":true,"oa_status":"hybrid","oa_url":"https://doi.org/10.1109/mc.2026.3652916","any_repository_has_fulltext":true},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5122486581","display_name":"Xiaoyu Ma","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Xiaoyu Ma","raw_affiliation_strings":["Google DeepMind, Mountain View, CA, USA"],"raw_orcid":"https://orcid.org/0009-0004-8066-6044","affiliations":[{"raw_affiliation_string":"Google DeepMind, Mountain View, CA, USA","institution_ids":["https://openalex.org/I1291425158"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5122691993","display_name":"David Patterson","orcid":null},"institutions":[{"id":"https://openalex.org/I1291425158","display_name":"Google (United States)","ror":"https://ror.org/00njsd438","country_code":"US","type":"company","lineage":["https://openalex.org/I1291425158","https://openalex.org/I4210128969"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"David Patterson","raw_affiliation_strings":["Google DeepMind, Mountain View, CA, USA"],"raw_orcid":"https://orcid.org/0000-0002-0429-1015","affiliations":[{"raw_affiliation_string":"Google DeepMind, Mountain View, CA, USA","institution_ids":["https://openalex.org/I1291425158"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I1291425158"],"apc_list":null,"apc_paid":null,"fwci":6.8132,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.92303937,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":"59","issue":"5","first_page":"55","last_page":"64"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10036","display_name":"Advanced Neural Network Applications","score":0.26269999146461487,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.26269999146461487,"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/T11948","display_name":"Machine Learning in Materials Science","score":0.14169999957084656,"subfield":{"id":"https://openalex.org/subfields/2505","display_name":"Materials Chemistry"},"field":{"id":"https://openalex.org/fields/25","display_name":"Materials 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":0.1128000020980835,"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/inference","display_name":"Inference","score":0.682200014591217},{"id":"https://openalex.org/keywords/language-model","display_name":"Language model","score":0.5814999938011169},{"id":"https://openalex.org/keywords/memory-bandwidth","display_name":"Memory bandwidth","score":0.527999997138977},{"id":"https://openalex.org/keywords/interconnection","display_name":"Interconnection","score":0.4595000147819519},{"id":"https://openalex.org/keywords/speedup","display_name":"Speedup","score":0.44690001010894775},{"id":"https://openalex.org/keywords/focus","display_name":"Focus (optics)","score":0.4465000033378601},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.41929998993873596},{"id":"https://openalex.org/keywords/mobile-device","display_name":"Mobile device","score":0.41100001335144043}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7886000275611877},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.682200014591217},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.5814999938011169},{"id":"https://openalex.org/C188045654","wikidata":"https://www.wikidata.org/wiki/Q17148339","display_name":"Memory bandwidth","level":2,"score":0.527999997138977},{"id":"https://openalex.org/C123745756","wikidata":"https://www.wikidata.org/wiki/Q1665949","display_name":"Interconnection","level":2,"score":0.4595000147819519},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.44749999046325684},{"id":"https://openalex.org/C68339613","wikidata":"https://www.wikidata.org/wiki/Q1549489","display_name":"Speedup","level":2,"score":0.44690001010894775},{"id":"https://openalex.org/C192209626","wikidata":"https://www.wikidata.org/wiki/Q190909","display_name":"Focus (optics)","level":2,"score":0.4465000033378601},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.41929998993873596},{"id":"https://openalex.org/C186967261","wikidata":"https://www.wikidata.org/wiki/Q5082128","display_name":"Mobile device","level":2,"score":0.41100001335144043},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.40059998631477356},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.38530001044273376},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.31119999289512634},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3021000027656555},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3018999993801117},{"id":"https://openalex.org/C2777472644","wikidata":"https://www.wikidata.org/wiki/Q16968992","display_name":"Approximate inference","level":3,"score":0.3012999892234802},{"id":"https://openalex.org/C2779602883","wikidata":"https://www.wikidata.org/wiki/Q15544750","display_name":"Memory architecture","level":2,"score":0.30079999566078186},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.29829999804496765},{"id":"https://openalex.org/C2776531357","wikidata":"https://www.wikidata.org/wiki/Q174077","display_name":"Flash memory","level":2,"score":0.2953999936580658},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.2953000068664551},{"id":"https://openalex.org/C67186912","wikidata":"https://www.wikidata.org/wiki/Q367664","display_name":"Data modeling","level":2,"score":0.28690001368522644},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.2784000039100647},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.27649998664855957},{"id":"https://openalex.org/C159877910","wikidata":"https://www.wikidata.org/wiki/Q2202883","display_name":"Autoregressive model","level":2,"score":0.27090001106262207},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.27059999108314514},{"id":"https://openalex.org/C179603123","wikidata":"https://www.wikidata.org/wiki/Q1941921","display_name":"Modeling language","level":3,"score":0.26019999384880066}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1109/mc.2026.3652916","is_oa":true,"landing_page_url":"https://doi.org/10.1109/mc.2026.3652916","pdf_url":null,"source":{"id":"https://openalex.org/S178916657","display_name":"Computer","issn_l":"0018-9162","issn":["0018-9162","1558-0814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer","raw_type":"journal-article"},{"id":"pmh:oai:arXiv.org:2601.05047","is_oa":true,"landing_page_url":"http://arxiv.org/abs/2601.05047","pdf_url":"https://arxiv.org/pdf/2601.05047","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"text"},{"id":"pmh:doi:10.48550/arxiv.2601.05047","is_oa":true,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4406922384","display_name":"Open MIND","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":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"Article"}],"best_oa_location":{"id":"doi:10.1109/mc.2026.3652916","is_oa":true,"landing_page_url":"https://doi.org/10.1109/mc.2026.3652916","pdf_url":null,"source":{"id":"https://openalex.org/S178916657","display_name":"Computer","issn_l":"0018-9162","issn":["0018-9162","1558-0814"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310320439","host_organization_name":"IEEE Computer Society","host_organization_lineage":["https://openalex.org/P4310320439","https://openalex.org/P4310319808"],"host_organization_lineage_names":["IEEE Computer Society","Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Computer","raw_type":"journal-article"},"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/4","display_name":"Quality Education","score":0.43106168508529663}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"Large":[0],"Language":[1],"Model":[2],"(LLM)":[3],"inference":[4,18],"is":[5,76],"hard.":[6],"The":[7],"autoregressive":[8],"Decode":[9],"phase":[10],"of":[11],"the":[12,28],"underlying":[13],"Transformer":[14],"model":[15],"makes":[16],"LLM":[17],"fundamentally":[19],"different":[20],"from":[21],"training.":[22],"Exacerbated":[23],"by":[24],"recent":[25],"AI":[26],"trends,":[27],"primary":[29],"challenges":[30],"are":[31],"memory":[32,53,65],"and":[33,59,67],"interconnect":[34,69],"rather":[35],"than":[36],"compute.":[37],"To":[38],"address":[39],"these":[40],"challenges,":[41],"we":[42,79],"highlight":[43],"four":[44],"architecture":[45],"research":[46],"opportunities:":[47],"High":[48],"Bandwidth":[49],"Flash":[50],"for":[51,63,84],"10X":[52],"capacity":[54],"with":[55],"HBM-like":[56],"bandwidth;":[57,66],"Processing-Near-Memory":[58],"3D":[60],"memory-logic":[61],"stacking":[62],"high":[64],"low-latency":[68],"to":[70],"speedup":[71],"communication.":[72],"While":[73],"our":[74],"focus":[75],"datacenter":[77],"AI,":[78],"also":[80],"review":[81],"their":[82],"applicability":[83],"mobile":[85],"devices.":[86]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-07-28T07:46:37.118299","created_date":"2026-01-10T00:00:00"}