{"id":"https://openalex.org/W4393168980","doi":"https://doi.org/10.1109/asp-dac58780.2024.10473968","title":"PRIMATE: Processing in Memory Acceleration for Dynamic Token-pruning Transformers","display_name":"PRIMATE: Processing in Memory Acceleration for Dynamic Token-pruning Transformers","publication_year":2024,"publication_date":"2024-01-22","ids":{"openalex":"https://openalex.org/W4393168980","doi":"https://doi.org/10.1109/asp-dac58780.2024.10473968"},"language":"en","primary_location":{"id":"doi:10.1109/asp-dac58780.2024.10473968","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asp-dac58780.2024.10473968","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 29th Asia and South Pacific Design Automation Conference (ASP-DAC)","raw_type":"proceedings-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/A5108053345","display_name":"Yue Pan","orcid":"https://orcid.org/0000-0003-4572-8882"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California, San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":true,"raw_author_name":"Yue Pan","raw_affiliation_strings":["University of California San Diego,La Jolla,CA,United States"],"affiliations":[{"raw_affiliation_string":"University of California San Diego,La Jolla,CA,United States","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5036778557","display_name":"Minxuan Zhou","orcid":"https://orcid.org/0000-0002-5523-7270"},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California, San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Minxuan Zhou","raw_affiliation_strings":["University of California San Diego,La Jolla,CA,United States"],"affiliations":[{"raw_affiliation_string":"University of California San Diego,La Jolla,CA,United States","institution_ids":["https://openalex.org/I36258959"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5051371646","display_name":"Chonghan Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I130769515","display_name":"Pennsylvania State University","ror":"https://ror.org/04p491231","country_code":"US","type":"education","lineage":["https://openalex.org/I130769515"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Chonghan Lee","raw_affiliation_strings":["Pennsylvania State University,State College,PA,United States"],"affiliations":[{"raw_affiliation_string":"Pennsylvania State University,State College,PA,United States","institution_ids":["https://openalex.org/I130769515"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101426729","display_name":"Zheyu Li","orcid":"https://orcid.org/0009-0005-4117-807X"},"institutions":[{"id":"https://openalex.org/I130769515","display_name":"Pennsylvania State University","ror":"https://ror.org/04p491231","country_code":"US","type":"education","lineage":["https://openalex.org/I130769515"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Zheyu Li","raw_affiliation_strings":["Pennsylvania State University,State College,PA,United States"],"affiliations":[{"raw_affiliation_string":"Pennsylvania State University,State College,PA,United States","institution_ids":["https://openalex.org/I130769515"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5011347197","display_name":"Rishika Kushwah","orcid":null},"institutions":[{"id":"https://openalex.org/I130769515","display_name":"Pennsylvania State University","ror":"https://ror.org/04p491231","country_code":"US","type":"education","lineage":["https://openalex.org/I130769515"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Rishika Kushwah","raw_affiliation_strings":["Pennsylvania State University,State College,PA,United States"],"affiliations":[{"raw_affiliation_string":"Pennsylvania State University,State College,PA,United States","institution_ids":["https://openalex.org/I130769515"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101919131","display_name":"Vijaykrishnan Narayanan","orcid":"https://orcid.org/0000-0001-6266-6068"},"institutions":[{"id":"https://openalex.org/I130769515","display_name":"Pennsylvania State University","ror":"https://ror.org/04p491231","country_code":"US","type":"education","lineage":["https://openalex.org/I130769515"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Vijaykrishnan Narayanan","raw_affiliation_strings":["Pennsylvania State University,State College,PA,United States"],"affiliations":[{"raw_affiliation_string":"Pennsylvania State University,State College,PA,United States","institution_ids":["https://openalex.org/I130769515"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5112841571","display_name":"Tajana Rosing","orcid":null},"institutions":[{"id":"https://openalex.org/I36258959","display_name":"University of California, San Diego","ror":"https://ror.org/0168r3w48","country_code":"US","type":"education","lineage":["https://openalex.org/I36258959"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Tajana Rosing","raw_affiliation_strings":["University of California San Diego,La Jolla,CA,United States"],"affiliations":[{"raw_affiliation_string":"University of California San Diego,La Jolla,CA,United States","institution_ids":["https://openalex.org/I36258959"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5108053345"],"corresponding_institution_ids":["https://openalex.org/I36258959"],"apc_list":null,"apc_paid":null,"fwci":1.0396,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.72360526,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":90,"max":95},"biblio":{"volume":null,"issue":null,"first_page":"557","last_page":"563"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9656999707221985,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9656999707221985,"subfield":{"id":"https://openalex.org/subfields/1708","display_name":"Hardware and Architecture"},"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/T11181","display_name":"Advanced Data Storage Technologies","score":0.9495999813079834,"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/primate","display_name":"Primate","score":0.8015715479850769},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.6323936581611633},{"id":"https://openalex.org/keywords/acceleration","display_name":"Acceleration","score":0.5453556776046753},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5442363023757935},{"id":"https://openalex.org/keywords/neuroscience","display_name":"Neuroscience","score":0.3622238039970398},{"id":"https://openalex.org/keywords/biology","display_name":"Biology","score":0.142675518989563},{"id":"https://openalex.org/keywords/computer-network","display_name":"Computer network","score":0.06403794884681702},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.06101146340370178}],"concepts":[{"id":"https://openalex.org/C2780509455","wikidata":"https://www.wikidata.org/wiki/Q7380","display_name":"Primate","level":2,"score":0.8015715479850769},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6323936581611633},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.5453556776046753},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5442363023757935},{"id":"https://openalex.org/C169760540","wikidata":"https://www.wikidata.org/wiki/Q207011","display_name":"Neuroscience","level":1,"score":0.3622238039970398},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.142675518989563},{"id":"https://openalex.org/C31258907","wikidata":"https://www.wikidata.org/wiki/Q1301371","display_name":"Computer network","level":1,"score":0.06403794884681702},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.06101146340370178},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/asp-dac58780.2024.10473968","is_oa":false,"landing_page_url":"https://doi.org/10.1109/asp-dac58780.2024.10473968","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2024 29th Asia and South Pacific Design Automation Conference (ASP-DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","score":0.7200000286102295,"display_name":"Affordable and clean energy"}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":31,"referenced_works":["https://openalex.org/W1797268635","https://openalex.org/W2034861439","https://openalex.org/W2129991978","https://openalex.org/W2170257519","https://openalex.org/W2251939518","https://openalex.org/W2472334012","https://openalex.org/W2761132374","https://openalex.org/W2765234579","https://openalex.org/W2896457183","https://openalex.org/W2949989598","https://openalex.org/W2955041501","https://openalex.org/W2976137532","https://openalex.org/W3016166938","https://openalex.org/W3017024317","https://openalex.org/W3094502228","https://openalex.org/W3100710793","https://openalex.org/W3100985894","https://openalex.org/W3134274954","https://openalex.org/W3155004489","https://openalex.org/W3159727696","https://openalex.org/W3168124404","https://openalex.org/W3192478068","https://openalex.org/W4280496502","https://openalex.org/W4308083827","https://openalex.org/W4312849330","https://openalex.org/W4379115947","https://openalex.org/W4381233128","https://openalex.org/W6691459498","https://openalex.org/W6755207826","https://openalex.org/W6784333009","https://openalex.org/W6796494063"],"related_works":["https://openalex.org/W2748952813","https://openalex.org/W4388335561","https://openalex.org/W2970530566","https://openalex.org/W2967478618","https://openalex.org/W4385009901","https://openalex.org/W4385572700","https://openalex.org/W4307309205","https://openalex.org/W4288261899","https://openalex.org/W2997152889","https://openalex.org/W4210509889"],"abstract_inverted_index":{"Attention-based":[0],"models":[1,44],"such":[2],"as":[3],"Transformers":[4],"represent":[5],"the":[6,9,24,63,76,92,102,177],"state":[7],"of":[8,78,94],"art":[10],"for":[11,65,154,182],"various":[12],"machine":[13],"learning":[14],"(ML)":[15],"tasks.":[16],"Their":[17],"superior":[18],"performance":[19],"is":[20,37],"often":[21],"overshadowed":[22],"by":[23],"substantial":[25],"memory":[26,56,103],"requirements":[27],"and":[28,54,101,139,150,157,170],"low":[29,50],"data":[30,51],"reuse":[31],"opportunities.":[32],"Processing":[33],"in":[34,163,167],"Memory":[35,124],"(PIM)":[36],"a":[38,115,146],"promising":[39],"solution":[40],"to":[41,46,84,131,134,176],"accelerate":[42],"Transformer":[43,136],"due":[45],"its":[47],"massive":[48],"parallelism,":[49],"movement":[52],"costs,":[53],"high":[55],"bandwidth":[57],"utilization.":[58],"Existing":[59],"PIM":[60,118,180],"accelerators":[61],"lack":[62,93],"support":[64],"algorithmic":[66],"optimizations":[67],"like":[68],"dynamic":[69,86],"token":[70,87,98],"pruning":[71,88],"that":[72],"can":[73],"significantly":[74],"improve":[75],"efficiency":[77,174],"Transformers.":[79,183],"We":[80,126],"identify":[81],"two":[82],"challenges":[83],"enabling":[85],"on":[89,121],"PIM-based":[90],"architectures:":[91],"an":[95,151],"in-memory":[96],"top-k":[97,140],"selection":[99],"mechanism":[100],"underutilization":[104],"problem":[105],"from":[106],"pruning.":[107],"To":[108],"address":[109],"these":[110],"challenges,":[111],"we":[112,144],"propose":[113],"PRIMATE,":[114],"software-hardware":[116],"co-design":[117],"framework":[119,153],"based":[120],"High":[122],"Bandwidth":[123],"(HBM).":[125],"initiate":[127],"minor":[128],"hardware":[129],"modifications":[130],"conventional":[132],"HBM":[133],"enable":[135],"model":[137],"computation":[138],"selection.":[141],"For":[142],"software,":[143],"introduce":[145],"pipelined":[147],"mapping":[148],"scheme":[149],"optimization":[152],"maximum":[155],"throughput":[156],"efficiency.":[158],"PRIMATE":[159],"achieves":[160],"$30.6\\times$":[161],"improvement":[162,166],"throughput,":[164],"$29.5\\times$":[165],"space":[168],"efficiency,":[169],"$4.3\\times$":[171],"better":[172],"energy":[173],"compared":[175],"current":[178],"state-of-the-art":[179],"accelerator":[181]},"counts_by_year":[{"year":2025,"cited_by_count":1},{"year":2024,"cited_by_count":1}],"updated_date":"2025-11-06T03:46:38.306776","created_date":"2025-10-10T00:00:00"}
