{"id":"https://openalex.org/W4411358592","doi":"https://doi.org/10.1109/tcsii.2025.3580135","title":"KV-Cache Oriented Query-Aware Sparse Attention Accelerator With Cross-Stage Precision-Configurable Digital CIM","display_name":"KV-Cache Oriented Query-Aware Sparse Attention Accelerator With Cross-Stage Precision-Configurable Digital CIM","publication_year":2025,"publication_date":"2025-06-17","ids":{"openalex":"https://openalex.org/W4411358592","doi":"https://doi.org/10.1109/tcsii.2025.3580135"},"language":"en","primary_location":{"id":"doi:10.1109/tcsii.2025.3580135","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsii.2025.3580135","pdf_url":null,"source":{"id":"https://openalex.org/S93916849","display_name":"IEEE Transactions on Circuits & Systems II Express Briefs","issn_l":"1549-7747","issn":["1549-7747","1558-3791"],"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 Transactions on Circuits and Systems II: Express Briefs","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/A5088633393","display_name":"Yang Zhang","orcid":"https://orcid.org/0009-0005-1455-5749"},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Yang Zhang","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0005-1455-5749","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109689226","display_name":"Xilong Kang","orcid":null},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xilong Kang","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0002-3853-2398","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Weixuan Wang","orcid":"https://orcid.org/0009-0008-4835-5720"},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Weixuan Wang","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0008-4835-5720","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101285366","display_name":"Yizhi Ding","orcid":null},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yizhi Ding","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5070617557","display_name":"Lizheng Ren","orcid":null},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lizheng Ren","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Yiran Zhang","orcid":"https://orcid.org/0009-0007-7903-4623"},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yiran Zhang","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0007-7903-4623","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":null,"display_name":"Ruiqi Tan","orcid":"https://orcid.org/0009-0001-0148-396X"},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Ruiqi Tan","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0001-0148-396X","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100670065","display_name":"Zhen Wang","orcid":"https://orcid.org/0000-0002-7821-2252"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Zhen Wang","raw_affiliation_strings":["Nanjing Prochip Electronic Technology Co. Ltd., Nanjing, China","Nanjing Prochip Electronic Technology Co. Ltd, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-7821-2252","affiliations":[{"raw_affiliation_string":"Nanjing Prochip Electronic Technology Co. Ltd., Nanjing, China","institution_ids":[]},{"raw_affiliation_string":"Nanjing Prochip Electronic Technology Co. Ltd, Nanjing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5079498326","display_name":"Hao Cai","orcid":"https://orcid.org/0000-0001-9794-8049"},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Hao Cai","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0001-9794-8049","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100461621","display_name":"Bo Liu","orcid":"https://orcid.org/0000-0002-0894-1054"},"institutions":[{"id":"https://openalex.org/I76569877","display_name":"Southeast University","ror":"https://ror.org/04ct4d772","country_code":"CN","type":"education","lineage":["https://openalex.org/I76569877"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Bo Liu","raw_affiliation_strings":["School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-0894-1054","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":10,"corresponding_author_ids":["https://openalex.org/A5088633393"],"corresponding_institution_ids":["https://openalex.org/I76569877"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.12431877,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"72","issue":"8","first_page":"1073","last_page":"1077"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9977999925613403,"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":0.9977999925613403,"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/T10715","display_name":"Distributed and Parallel Computing Systems","score":0.993399977684021,"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"}},{"id":"https://openalex.org/T10904","display_name":"Embedded Systems Design Techniques","score":0.9750000238418579,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.7620664834976196},{"id":"https://openalex.org/keywords/stage","display_name":"Stage (stratigraphy)","score":0.5493839979171753},{"id":"https://openalex.org/keywords/cache","display_name":"Cache","score":0.5059563517570496},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.43432891368865967},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.3788365125656128}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7620664834976196},{"id":"https://openalex.org/C146357865","wikidata":"https://www.wikidata.org/wiki/Q1123245","display_name":"Stage (stratigraphy)","level":2,"score":0.5493839979171753},{"id":"https://openalex.org/C115537543","wikidata":"https://www.wikidata.org/wiki/Q165596","display_name":"Cache","level":2,"score":0.5059563517570496},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.43432891368865967},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.3788365125656128},{"id":"https://openalex.org/C151730666","wikidata":"https://www.wikidata.org/wiki/Q7205","display_name":"Paleontology","level":1,"score":0.0},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tcsii.2025.3580135","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tcsii.2025.3580135","pdf_url":null,"source":{"id":"https://openalex.org/S93916849","display_name":"IEEE Transactions on Circuits & Systems II Express Briefs","issn_l":"1549-7747","issn":["1549-7747","1558-3791"],"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 Transactions on Circuits and Systems II: Express Briefs","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G120510889","display_name":null,"funder_award_id":"2023YFB4403103","funder_id":"https://openalex.org/F4320335777","funder_display_name":"National Key Research and Development Program of China"}],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":24,"referenced_works":["https://openalex.org/W4231563577","https://openalex.org/W4360606423","https://openalex.org/W4360606782","https://openalex.org/W4380881077","https://openalex.org/W4384918448","https://openalex.org/W4385245566","https://openalex.org/W4386763770","https://openalex.org/W4386763875","https://openalex.org/W4387042316","https://openalex.org/W4387321091","https://openalex.org/W4392746069","https://openalex.org/W4393161108","https://openalex.org/W4399794499","https://openalex.org/W4401210315","https://openalex.org/W4404132796","https://openalex.org/W4404133659","https://openalex.org/W4408163921","https://openalex.org/W6778883912","https://openalex.org/W6838322825","https://openalex.org/W6846659131","https://openalex.org/W6848451824","https://openalex.org/W6854866820","https://openalex.org/W6869510815","https://openalex.org/W6873851262"],"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":{"This":[0],"work":[1],"proposes":[2],"KV-CIM,":[3],"a":[4],"KV-Cache":[5,39,50],"oriented":[6],"Digital":[7],"Compute-In-Memory":[8],"(DCIM)":[9],"sparse":[10],"attention":[11],"accelerator,":[12],"to":[13,84,104],"address":[14],"computational":[15,58],"and":[16,37,48,82,118],"memory":[17,123],"bottlenecks":[18],"in":[19,77,100,128],"autoregressive":[20],"inference":[21],"for":[22,40],"large":[23],"language":[24],"models.":[25],"Key":[26],"innovations":[27],"include:":[28],"a)":[29],"A":[30,67,91],"query-aware":[31],"pre-compute":[32,45],"architecture,":[33],"which":[34,75],"dynamically":[35],"selects":[36],"accesses":[38],"critical":[41],"tokens":[42],"at":[43,60,80,88],"the":[44,61,96,112],"stage":[46,64],"(Stage1)":[47],"deploys":[49],"segmentally":[51],"on":[52],"memory-constrained":[53],"edge":[54],"devices":[55],"while":[56],"maintaining":[57],"accuracy":[59],"formal":[62],"computation":[63],"(Stage2);":[65],"b)":[66],"cross-stage":[68],"DCIM":[69],"macro":[70],"featuring":[71],"precision-configurable":[72],"adder":[73],"trees,":[74],"works":[76],"approximate":[78],"mode":[79,87],"Stage1":[81],"changes":[83],"full":[85],"precision":[86],"Stage2;":[89],"c)":[90],"query-stationary":[92],"dataflow":[93],"that":[94],"retains":[95],"current":[97],"query":[98],"tensors":[99],"q-CIM":[101],"across":[102],"stages":[103],"eliminate":[105],"data":[106],"movement.":[107],"Under":[108],"28-nm":[109],"CMOS":[110],"technology,":[111],"proposed":[113],"KV-CIM":[114],"achieves":[115],"35.16":[116],"TOPS/W":[117],"82%":[119],"reduction":[120],"of":[121],"external":[122],"access":[124],"with":[125],"negligible":[126],"degradation":[127],"LLaMA2":[129],"expressiveness.":[130]},"counts_by_year":[],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}
