{"id":"https://openalex.org/W7092181620","doi":"https://doi.org/10.1145/3725843.3756062","title":"LongSight: Compute-Enabled Memory to Accelerate Large-Context LLMs via Sparse Attention","display_name":"LongSight: Compute-Enabled Memory to Accelerate Large-Context LLMs via Sparse Attention","publication_year":2025,"publication_date":"2025-10-17","ids":{"openalex":"https://openalex.org/W7092181620","doi":"https://doi.org/10.1145/3725843.3756062"},"language":null,"primary_location":{"id":"doi:10.1145/3725843.3756062","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725843.3756062","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3725843.3756062","source":null,"license":null,"license_id":null,"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":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3725843.3756062","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Derrick Quinn","orcid":"https://orcid.org/0009-0000-5862-6565"},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Derrick Quinn","raw_affiliation_strings":["Cornell University, Ithaca, USA"],"raw_orcid":"https://orcid.org/0009-0000-5862-6565","affiliations":[{"raw_affiliation_string":"Cornell University, Ithaca, USA","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"middle","author":{"id":null,"display_name":"E. Ezgi Y\u00fccel","orcid":"https://orcid.org/0009-0000-0460-8230"},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"E. Ezgi Y\u00fccel","raw_affiliation_strings":["Cornell University, Ithaca, USA"],"raw_orcid":"https://orcid.org/0009-0000-0460-8230","affiliations":[{"raw_affiliation_string":"Cornell University, Ithaca, USA","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jinkwon Kim","orcid":"https://orcid.org/0000-0002-1744-1393"},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jinkwon Kim","raw_affiliation_strings":["Cornell University, Ithaca, USA"],"raw_orcid":"https://orcid.org/0000-0002-1744-1393","affiliations":[{"raw_affiliation_string":"Cornell University, Ithaca, USA","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Jos\u00e9 F. Mart\u00ednez","orcid":"https://orcid.org/0000-0001-5451-5681"},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jos\u00e9 F. Mart\u00ednez","raw_affiliation_strings":["Cornell University, Ithaca, USA"],"raw_orcid":"https://orcid.org/0000-0001-5451-5681","affiliations":[{"raw_affiliation_string":"Cornell University, Ithaca, USA","institution_ids":["https://openalex.org/I205783295"]}]},{"author_position":"last","author":{"id":null,"display_name":"Mohammad Alian","orcid":"https://orcid.org/0000-0002-4622-2181"},"institutions":[{"id":"https://openalex.org/I205783295","display_name":"Cornell University","ror":"https://ror.org/05bnh6r87","country_code":"US","type":"education","lineage":["https://openalex.org/I205783295"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Mohammad Alian","raw_affiliation_strings":["Cornell University, Ithaca, USA"],"raw_orcid":"https://orcid.org/0000-0002-4622-2181","affiliations":[{"raw_affiliation_string":"Cornell University, Ithaca, USA","institution_ids":["https://openalex.org/I205783295"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I205783295"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.57092591,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"34","last_page":"48"},"is_retracted":false,"is_paratext":false,"is_xpac":true,"primary_topic":{"id":"https://openalex.org/T10028","display_name":"Topic Modeling","score":0.11680000275373459,"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/T10028","display_name":"Topic Modeling","score":0.11680000275373459,"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/T10036","display_name":"Advanced Neural Network Applications","score":0.11320000141859055,"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/T10181","display_name":"Natural Language Processing Techniques","score":0.1103999987244606,"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/key","display_name":"Key (lock)","score":0.3073999881744385},{"id":"https://openalex.org/keywords/feature","display_name":"Feature (linguistics)","score":0.26019999384880066},{"id":"https://openalex.org/keywords/noise","display_name":"Noise (video)","score":0.23929999768733978},{"id":"https://openalex.org/keywords/perception","display_name":"Perception","score":0.2379000037908554}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5666000247001648},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3779999911785126},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.3073999881744385},{"id":"https://openalex.org/C180747234","wikidata":"https://www.wikidata.org/wiki/Q23373","display_name":"Cognitive psychology","level":1,"score":0.27559998631477356},{"id":"https://openalex.org/C38652104","wikidata":"https://www.wikidata.org/wiki/Q3510521","display_name":"Computer security","level":1,"score":0.26910001039505005},{"id":"https://openalex.org/C2776401178","wikidata":"https://www.wikidata.org/wiki/Q12050496","display_name":"Feature (linguistics)","level":2,"score":0.26019999384880066},{"id":"https://openalex.org/C15744967","wikidata":"https://www.wikidata.org/wiki/Q9418","display_name":"Psychology","level":0,"score":0.24459999799728394},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.23929999768733978},{"id":"https://openalex.org/C188147891","wikidata":"https://www.wikidata.org/wiki/Q147638","display_name":"Cognitive science","level":1,"score":0.2387000024318695},{"id":"https://openalex.org/C26760741","wikidata":"https://www.wikidata.org/wiki/Q160402","display_name":"Perception","level":2,"score":0.2379000037908554}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3725843.3756062","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725843.3756062","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3725843.3756062","source":null,"license":null,"license_id":null,"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"}],"best_oa_location":{"id":"doi:10.1145/3725843.3756062","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3725843.3756062","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3725843.3756062","source":null,"license":null,"license_id":null,"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/G1452086972","display_name":null,"funder_award_id":"JUMP 2.0","funder_id":"https://openalex.org/F4320306087","funder_display_name":"Semiconductor Research Corporation"},{"id":"https://openalex.org/G599137258","display_name":null,"funder_award_id":"JUMP 2.0","funder_id":"https://openalex.org/F4320332180","funder_display_name":"Defense Advanced Research Projects Agency"},{"id":"https://openalex.org/G7089484908","display_name":null,"funder_award_id":"CCF-2530337, CCF-2312741","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G7869976093","display_name":"Collaborative Research: SHF: Medium: A hardware-software co-design approach for high-performance in-memory analytic data processing","funder_award_id":"2312741","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"},{"id":"https://openalex.org/G911691874","display_name":"CAREER: Near-Memory Datacenter Network","funder_award_id":"2530337","funder_id":"https://openalex.org/F4320306076","funder_display_name":"National Science Foundation"}],"funders":[{"id":"https://openalex.org/F4320306076","display_name":"National Science Foundation","ror":"https://ror.org/021nxhr62"},{"id":"https://openalex.org/F4320306087","display_name":"Semiconductor Research Corporation","ror":"https://ror.org/047z4n946"},{"id":"https://openalex.org/F4320332180","display_name":"Defense Advanced Research Projects Agency","ror":"https://ror.org/02caytj08"}],"has_content":{"grobid_xml":true,"pdf":true},"content_urls":{"pdf":"https://content.openalex.org/works/W7092181620.pdf","grobid_xml":"https://content.openalex.org/works/W7092181620.grobid-xml"},"referenced_works_count":21,"referenced_works":["https://openalex.org/W2034861439","https://openalex.org/W2084363474","https://openalex.org/W2588191434","https://openalex.org/W3006586535","https://openalex.org/W3036079062","https://openalex.org/W3189166979","https://openalex.org/W4318541517","https://openalex.org/W4318541614","https://openalex.org/W4327930439","https://openalex.org/W4387294617","https://openalex.org/W4389518760","https://openalex.org/W4389524535","https://openalex.org/W4391940867","https://openalex.org/W4392427708","https://openalex.org/W4393407316","https://openalex.org/W4394998968","https://openalex.org/W4396601595","https://openalex.org/W4401211635","https://openalex.org/W4408902531","https://openalex.org/W4411471835","https://openalex.org/W6947991961"],"related_works":[],"abstract_inverted_index":{"Large":[0],"input":[1],"context":[2,18,122],"windows":[3],"in":[4,67],"transformer-based":[5],"LLMs":[6],"help":[7],"minimize":[8],"hallucinations":[9],"and":[10,14,87,111],"improve":[11],"output":[12],"accuracy":[13],"personalization.However,":[15],"as":[16],"the":[17,21,40,47,92],"window":[19],"grows,":[20],"attention":[22,66],"phase":[23],"increasingly":[24],"dominates":[25],"execution":[26],"time.Key-Value":[27],"(KV)":[28],"caching":[29],"alleviates":[30],"part":[31],"of":[32,49,94,101,124],"this":[33,55],"cost":[34],"by":[35],"avoiding":[36],"redundant":[37],"computation,":[38],"but":[39],"KV":[41,84],"cache":[42,85],"itself":[43],"can":[44,119],"quickly":[45],"exceed":[46],"capacity":[48],"today's":[50],"GPU":[51,110],"high-bandwidth":[52],"memory":[53,74,116],"(HBM).In":[54],"work,":[56],"we":[57],"present":[58],"LongSight,":[59],"an":[60],"algorithm-hardware":[61],"co-design":[62],"framework":[63],"for":[64,78,130],"accelerating":[65],"large-context":[68],"scenarios.LongSight":[69],"leverages":[70],"a":[71,108,112],"compute-enabled":[72],"CXL":[73,115],"device,":[75],"originally":[76],"designed":[77],"dense":[79],"retrieval":[80],"acceleration,":[81],"to":[82,99,126],"offload":[83],"storage":[86],"retrieval.Therefore,":[88],"LongSight":[89,118],"effectively":[90],"elevates":[91],"value":[93],"relatively":[95],"low-cost":[96],"LPDDR":[97],"DRAM":[98],"that":[100],"high-end":[102],"HBM.We":[103],"demonstrate":[104],"that,":[105],"with":[106],"just":[107],"single":[109,113],"computeenabled":[114],"expander,":[117],"efficiently":[120],"support":[121],"lengths":[123],"up":[125],"1":[127],"million":[128],"tokens":[129],"state-of-the-art":[131],"Llama":[132],"models.":[133]},"counts_by_year":[],"updated_date":"2026-07-31T08:31:51.225901","created_date":"2025-10-17T00:00:00"}