{"id":"https://openalex.org/W7161273637","doi":"https://doi.org/10.1109/lca.2026.3693796","title":"Towards Segmentation-Based Address Translation for LLM Inference","display_name":"Towards Segmentation-Based Address Translation for LLM Inference","publication_year":2026,"publication_date":"2026-01-01","ids":{"openalex":"https://openalex.org/W7161273637","doi":"https://doi.org/10.1109/lca.2026.3693796"},"language":null,"primary_location":{"id":"doi:10.1109/lca.2026.3693796","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lca.2026.3693796","pdf_url":null,"source":{"id":"https://openalex.org/S17643076","display_name":"IEEE Computer Architecture Letters","issn_l":"1556-6056","issn":["1556-6056","1556-6064","2473-2575"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Computer Architecture Letters","raw_type":"journal-article"},"type":"article","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/A5067284583","display_name":"Youngjoon Cheon","orcid":null},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Youngjoon Cheon","raw_affiliation_strings":["Korea University, Seoul, South Korea"],"raw_orcid":"https://orcid.org/0009-0002-0022-8876","affiliations":[{"raw_affiliation_string":"Korea University, Seoul, South Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5136191829","display_name":"Yunho Oh","orcid":"https://orcid.org/0000-0001-6442-3705"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Yunho Oh","raw_affiliation_strings":["Korea University, Seoul, South Korea"],"raw_orcid":"https://orcid.org/0000-0001-6442-3705","affiliations":[{"raw_affiliation_string":"Korea University, Seoul, South Korea","institution_ids":["https://openalex.org/I197347611"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5037825380","display_name":"Jeongseob Ahn","orcid":"https://orcid.org/0000-0002-4503-8294"},"institutions":[{"id":"https://openalex.org/I197347611","display_name":"Korea University","ror":"https://ror.org/047dqcg40","country_code":"KR","type":"education","lineage":["https://openalex.org/I197347611"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jeongseob Ahn","raw_affiliation_strings":["Korea University, Seoul, South Korea"],"raw_orcid":"https://orcid.org/0000-0002-4503-8294","affiliations":[{"raw_affiliation_string":"Korea University, Seoul, South Korea","institution_ids":["https://openalex.org/I197347611"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.77166198,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"25","issue":"1","first_page":"190","last_page":"193"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10181","display_name":"Natural Language Processing Techniques","score":0.7421000003814697,"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":0.7421000003814697,"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/T10215","display_name":"Semantic Web and Ontologies","score":0.0738999992609024,"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":0.05649999901652336,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5547999739646912},{"id":"https://openalex.org/keywords/translation","display_name":"Translation (biology)","score":0.4758000075817108},{"id":"https://openalex.org/keywords/semantics","display_name":"Semantics (computer science)","score":0.33399999141693115},{"id":"https://openalex.org/keywords/machine-translation","display_name":"Machine translation","score":0.3098999857902527},{"id":"https://openalex.org/keywords/process","display_name":"Process (computing)","score":0.2718000113964081}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.8233000040054321},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5547999739646912},{"id":"https://openalex.org/C204321447","wikidata":"https://www.wikidata.org/wiki/Q30642","display_name":"Natural language processing","level":1,"score":0.49160000681877136},{"id":"https://openalex.org/C149364088","wikidata":"https://www.wikidata.org/wiki/Q185917","display_name":"Translation (biology)","level":4,"score":0.4758000075817108},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4754999876022339},{"id":"https://openalex.org/C199360897","wikidata":"https://www.wikidata.org/wiki/Q9143","display_name":"Programming language","level":1,"score":0.37400001287460327},{"id":"https://openalex.org/C184337299","wikidata":"https://www.wikidata.org/wiki/Q1437428","display_name":"Semantics (computer science)","level":2,"score":0.33399999141693115},{"id":"https://openalex.org/C203005215","wikidata":"https://www.wikidata.org/wiki/Q79798","display_name":"Machine translation","level":2,"score":0.3098999857902527},{"id":"https://openalex.org/C98045186","wikidata":"https://www.wikidata.org/wiki/Q205663","display_name":"Process (computing)","level":2,"score":0.2718000113964081},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.26669999957084656},{"id":"https://openalex.org/C195818886","wikidata":"https://www.wikidata.org/wiki/Q5421724","display_name":"Expressive power","level":2,"score":0.2526000142097473}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/lca.2026.3693796","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lca.2026.3693796","pdf_url":null,"source":{"id":"https://openalex.org/S17643076","display_name":"IEEE Computer Architecture Letters","issn_l":"1556-6056","issn":["1556-6056","1556-6064","2473-2575"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Computer Architecture Letters","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":0,"referenced_works":[],"related_works":[],"abstract_inverted_index":{"1As":[0],"large":[1],"language":[2],"models":[3],"(LLMs)":[4],"continue":[5],"to":[6,56,71,97],"scale":[7],"in":[8,34,105],"both":[9],"model":[10,32,69,77,103,146],"size":[11],"and":[12,81],"context":[13],"length,":[14],"the":[15,44,85,91,128,137],"efficiency":[16],"of":[17,87],"address":[18,66,93,121],"translation":[19,67,94,118,122,143],"has":[20],"become":[21],"a":[22,35,88,106,112],"critical":[23],"performance":[24],"bottleneck":[25],"during":[26],"inference.":[27],"Although":[28],"LLM":[29],"inference":[30],"accesses":[31],"weights":[33,70,78,104],"layer-wise":[36],"pattern":[37],"with":[38],"substantial":[39],"reuse":[40],"across":[41],"token":[42],"generations,":[43],"expanding":[45],"KV-cache":[46],"footprint":[47],"increasingly":[48],"contends":[49],"for":[50,68,84,145],"TLB":[51],"capacity,":[52],"leaving":[53],"fewer":[54],"entries":[55],"cache":[57],"weight":[58],"translations.":[59],"In":[60],"this":[61,98],"paper,":[62],"we":[63],"revisit":[64],"segmentation-based":[65,92],"bypass":[72],"conventional":[73],"TLB-based":[74],"translation.":[75],"Since":[76],"are":[79],"read-only":[80],"not":[82],"reclaimed":[83],"lifetime":[86],"serving":[89],"instance,":[90],"is":[95],"well-suited":[96],"access":[99],"pattern.":[100],"We":[101],"place":[102],"virtually":[107],"contiguous":[108,114],"region":[109],"backed":[110],"by":[111,134,140],"physically":[113],"memory,":[115],"enabling":[116],"segment-based":[117],"that":[119,127],"minimizes":[120],"overhead.":[123],"Our":[124],"evaluation":[125],"shows":[126],"proposed":[129],"design":[130],"improves":[131],"overall":[132],"IPC":[133],"2.51\u00d7":[135],"over":[136],"paging":[138],"baseline":[139],"achieving":[141],"near-zero":[142],"overhead":[144],"weights.":[147]},"counts_by_year":[],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2026-05-16T00:00:00"}