{"id":"https://openalex.org/W4410028046","doi":"https://doi.org/10.1109/lca.2025.3566692","title":"Fold-PIM: A Cost-Efficient LPDDR5-Based PIM for On-Device SLMs","display_name":"Fold-PIM: A Cost-Efficient LPDDR5-Based PIM for On-Device SLMs","publication_year":2025,"publication_date":"2025-01-01","ids":{"openalex":"https://openalex.org/W4410028046","doi":"https://doi.org/10.1109/lca.2025.3566692"},"language":"en","primary_location":{"id":"doi:10.1109/lca.2025.3566692","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lca.2025.3566692","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/A5105015451","display_name":"Kyoungho Jeun","orcid":null},"institutions":[{"id":"https://openalex.org/I191879574","display_name":"Inha University","ror":"https://ror.org/01easw929","country_code":"KR","type":"education","lineage":["https://openalex.org/I191879574"]}],"countries":["KR"],"is_corresponding":true,"raw_author_name":"Kyoungho Jeun","raw_affiliation_strings":["Inha University, Incheon, South Korea"],"affiliations":[{"raw_affiliation_string":"Inha University, Incheon, South Korea","institution_ids":["https://openalex.org/I191879574"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5106421113","display_name":"Hyeonu Kim","orcid":"https://orcid.org/0009-0008-5951-2541"},"institutions":[{"id":"https://openalex.org/I191879574","display_name":"Inha University","ror":"https://ror.org/01easw929","country_code":"KR","type":"education","lineage":["https://openalex.org/I191879574"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyeonu Kim","raw_affiliation_strings":["Inha University, Incheon, South Korea"],"affiliations":[{"raw_affiliation_string":"Inha University, Incheon, South Korea","institution_ids":["https://openalex.org/I191879574"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5035539518","display_name":"Eojin Lee","orcid":"https://orcid.org/0000-0003-2739-2924"},"institutions":[{"id":"https://openalex.org/I191879574","display_name":"Inha University","ror":"https://ror.org/01easw929","country_code":"KR","type":"education","lineage":["https://openalex.org/I191879574"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Eojin Lee","raw_affiliation_strings":["Inha University, Incheon, South Korea"],"affiliations":[{"raw_affiliation_string":"Inha University, Incheon, South Korea","institution_ids":["https://openalex.org/I191879574"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":3,"corresponding_author_ids":["https://openalex.org/A5105015451"],"corresponding_institution_ids":["https://openalex.org/I191879574"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.10695743,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"24","issue":"1","first_page":"185","last_page":"188"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T12371","display_name":"Electrical Contact Performance and Analysis","score":0.9775999784469604,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical 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/T12371","display_name":"Electrical Contact Performance and Analysis","score":0.9775999784469604,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical 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/T11527","display_name":"3D IC and TSV technologies","score":0.9771999716758728,"subfield":{"id":"https://openalex.org/subfields/2208","display_name":"Electrical and Electronic Engineering"},"field":{"id":"https://openalex.org/fields/22","display_name":"Engineering"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6987210512161255},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3938480317592621}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6987210512161255},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3938480317592621}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/lca.2025.3566692","is_oa":false,"landing_page_url":"https://doi.org/10.1109/lca.2025.3566692","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":[{"score":0.5699999928474426,"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W2982008795","https://openalex.org/W3146763006","https://openalex.org/W3147974890","https://openalex.org/W3189166979","https://openalex.org/W4297097348","https://openalex.org/W4362707004","https://openalex.org/W4393145094","https://openalex.org/W4393407316","https://openalex.org/W4405029226","https://openalex.org/W4411019820","https://openalex.org/W6778883912","https://openalex.org/W6861371496","https://openalex.org/W6875025663"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"The":[0],"increasing":[1],"demand":[2],"for":[3,16,30,82],"on-device":[4],"AI":[5],"applications":[6],"has":[7],"shifted":[8],"focus":[9],"to":[10,55,94,102,115,126],"Small":[11],"Language":[12],"Models":[13],"(SLMs)":[14],"optimized":[15],"mobile":[17],"environments.":[18],"However,":[19],"the":[20,127],"limited":[21],"memory":[22],"bandwidth":[23],"of":[24,40,118],"LPDDR5-based":[25,50],"systems":[26],"presents":[27],"significant":[28],"challenges":[29],"efficiently":[31],"executing":[32],"memory-bound":[33],"matrix-vector":[34],"multiplication":[35],"operations,":[36],"a":[37,61,90,99],"core":[38],"component":[39],"SLM":[41,123],"inference.":[42],"In":[43],"this":[44],"paper,":[45],"we":[46,75],"propose":[47],"Fold-PIM,":[48],"an":[49],"Processing-in-Memory":[51],"(PIM)":[52],"architecture":[53,64],"designed":[54],"address":[56],"these":[57],"challenges.":[58],"FoldPIM":[59],"features":[60],"shared":[62],"PU":[63],"that":[65,111],"leverages":[66],"subarray":[67],"level":[68],"parallelism":[69],"while":[70],"maintaining":[71],"cost":[72],"efficiency.":[73],"Additionally,":[74],"introduce":[76],"key":[77],"techniques,":[78],"including":[79],"adaptive":[80],"tiling":[81],"efficient":[83],"GEMV":[84],"operations":[85],"across":[86],"diverse":[87],"matrix":[88],"sizes,":[89],"transposed":[91],"data":[92],"layout":[93],"minimize":[95],"buffer":[96],"overhead,":[97],"and":[98],"tailored":[100],"protocol":[101],"reduce":[103],"vector":[104],"replacement":[105],"latency.":[106],"Our":[107],"evaluation":[108],"results":[109],"demonstrate":[110],"Fold-PIM":[112],"achieves":[113],"up":[114],"3.9\u00d7":[116],"speedup":[117],"token":[119],"generation":[120],"time":[121],"in":[122],"inference":[124],"compared":[125],"baseline":[128],"system":[129],"without":[130],"PIM.":[131]},"counts_by_year":[],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}