{"id":"https://openalex.org/W7166296795","doi":"https://doi.org/10.1145/3801487.3801814","title":"SABRE: A Compression-Aware BF16 Accelerator for Neuromorphic Attention","display_name":"SABRE: A Compression-Aware BF16 Accelerator for Neuromorphic Attention","publication_year":2026,"publication_date":"2026-05-19","ids":{"openalex":"https://openalex.org/W7166296795","doi":"https://doi.org/10.1145/3801487.3801814"},"language":null,"primary_location":{"id":"doi:10.1145/3801487.3801814","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3801487.3801814","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 23rd ACM International Conference on Computing Frontiers","raw_type":"proceedings-article"},"type":"article","indexed_in":["crossref"],"open_access":{"is_oa":true,"oa_status":"gold","oa_url":"https://doi.org/10.1145/3801487.3801814","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":"https://openalex.org/A5109524357","display_name":"Kunal Kishore","orcid":null},"institutions":[{"id":"https://openalex.org/I196608512","display_name":"PES University","ror":"https://ror.org/05m169e78","country_code":"IN","type":"education","lineage":["https://openalex.org/I196608512"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Kunal Kishore","raw_affiliation_strings":["Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India"],"raw_orcid":"https://orcid.org/0009-0001-2391-2710","affiliations":[{"raw_affiliation_string":"Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India","institution_ids":["https://openalex.org/I196608512"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139466335","display_name":"Manish Nagaraju","orcid":"https://orcid.org/0009-0009-6667-2129"},"institutions":[{"id":"https://openalex.org/I196608512","display_name":"PES University","ror":"https://ror.org/05m169e78","country_code":"IN","type":"education","lineage":["https://openalex.org/I196608512"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Manish Nagaraju","raw_affiliation_strings":["Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India"],"raw_orcid":"https://orcid.org/0009-0009-6667-2129","affiliations":[{"raw_affiliation_string":"Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India","institution_ids":["https://openalex.org/I196608512"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5135527552","display_name":"Anshul Raghavendra Katti","orcid":null},"institutions":[{"id":"https://openalex.org/I196608512","display_name":"PES University","ror":"https://ror.org/05m169e78","country_code":"IN","type":"education","lineage":["https://openalex.org/I196608512"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Anshul Raghavendra Katti","raw_affiliation_strings":["Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India"],"raw_orcid":"https://orcid.org/0009-0006-9224-8768","affiliations":[{"raw_affiliation_string":"Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India","institution_ids":["https://openalex.org/I196608512"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5139482464","display_name":"Nandan Venkataramana Bhat","orcid":"https://orcid.org/0009-0004-1684-597X"},"institutions":[{"id":"https://openalex.org/I196608512","display_name":"PES University","ror":"https://ror.org/05m169e78","country_code":"IN","type":"education","lineage":["https://openalex.org/I196608512"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Nandan Venkataramana Bhat","raw_affiliation_strings":["Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India"],"raw_orcid":"https://orcid.org/0009-0004-1684-597X","affiliations":[{"raw_affiliation_string":"Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India","institution_ids":["https://openalex.org/I196608512"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5074864897","display_name":"Pramod Udupa","orcid":"https://orcid.org/0000-0003-4943-3643"},"institutions":[{"id":"https://openalex.org/I4210146682","display_name":"Intel (India)","ror":"https://ror.org/04f2n1245","country_code":"IN","type":"company","lineage":["https://openalex.org/I1343180700","https://openalex.org/I4210146682"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Pramod Udupa","raw_affiliation_strings":["Intel, Bengaluru, Karnataka, India"],"raw_orcid":"https://orcid.org/0000-0003-4943-3643","affiliations":[{"raw_affiliation_string":"Intel, Bengaluru, Karnataka, India","institution_ids":["https://openalex.org/I4210146682"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5009362621","display_name":"Shikha Tripathi","orcid":"https://orcid.org/0000-0001-8123-5570"},"institutions":[{"id":"https://openalex.org/I196608512","display_name":"PES University","ror":"https://ror.org/05m169e78","country_code":"IN","type":"education","lineage":["https://openalex.org/I196608512"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Shikha Tripathi","raw_affiliation_strings":["Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India"],"raw_orcid":"https://orcid.org/0000-0001-8123-5570","affiliations":[{"raw_affiliation_string":"Department of Electronics and Communication, PES University, Bengaluru, Karnataka, India","institution_ids":["https://openalex.org/I196608512"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5072657959","display_name":"T. S. B. Sudarshan","orcid":"https://orcid.org/0000-0001-9032-7389"},"institutions":[{"id":"https://openalex.org/I4400573291","display_name":"Dayananda Sagar University","ror":"https://ror.org/033f7da12","country_code":"IN","type":"education","lineage":["https://openalex.org/I4400573291"]},{"id":"https://openalex.org/I8977528","display_name":"Dr. Hari Singh Gour University","ror":"https://ror.org/01xapxe37","country_code":"IN","type":"education","lineage":["https://openalex.org/I8977528"]}],"countries":["IN"],"is_corresponding":false,"raw_author_name":"Sudarshan TSB","raw_affiliation_strings":["Department of CSE, Dayananda Sagar University, Bengaluru, Karnataka, India"],"raw_orcid":"https://orcid.org/0000-0001-9032-7389","affiliations":[{"raw_affiliation_string":"Department of CSE, Dayananda Sagar University, Bengaluru, Karnataka, India","institution_ids":["https://openalex.org/I8977528","https://openalex.org/I4400573291"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":4,"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":0.88548267,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"188","last_page":"196"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.6877999901771545,"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"}},"topics":[{"id":"https://openalex.org/T10502","display_name":"Advanced Memory and Neural Computing","score":0.6877999901771545,"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"}},{"id":"https://openalex.org/T12808","display_name":"Ferroelectric and Negative Capacitance Devices","score":0.14190000295639038,"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"}},{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.05999999865889549,"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/neuromorphic-engineering","display_name":"Neuromorphic engineering","score":0.861299991607666},{"id":"https://openalex.org/keywords/quadratic-equation","display_name":"Quadratic equation","score":0.44449999928474426},{"id":"https://openalex.org/keywords/data-compression","display_name":"Data compression","score":0.41690000891685486},{"id":"https://openalex.org/keywords/kernel","display_name":"Kernel (algebra)","score":0.40950000286102295},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.40470001101493835},{"id":"https://openalex.org/keywords/compression","display_name":"Compression (physics)","score":0.3815999925136566},{"id":"https://openalex.org/keywords/limit","display_name":"Limit (mathematics)","score":0.37450000643730164},{"id":"https://openalex.org/keywords/singular-value-decomposition","display_name":"Singular value decomposition","score":0.3693999946117401},{"id":"https://openalex.org/keywords/latency","display_name":"Latency (audio)","score":0.367900013923645}],"concepts":[{"id":"https://openalex.org/C151927369","wikidata":"https://www.wikidata.org/wiki/Q1981312","display_name":"Neuromorphic engineering","level":3,"score":0.861299991607666},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7299000024795532},{"id":"https://openalex.org/C129844170","wikidata":"https://www.wikidata.org/wiki/Q41299","display_name":"Quadratic equation","level":2,"score":0.44449999928474426},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.41690000891685486},{"id":"https://openalex.org/C74193536","wikidata":"https://www.wikidata.org/wiki/Q574844","display_name":"Kernel (algebra)","level":2,"score":0.40950000286102295},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.40470001101493835},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.3837999999523163},{"id":"https://openalex.org/C180016635","wikidata":"https://www.wikidata.org/wiki/Q2712821","display_name":"Compression (physics)","level":2,"score":0.3815999925136566},{"id":"https://openalex.org/C151201525","wikidata":"https://www.wikidata.org/wiki/Q177239","display_name":"Limit (mathematics)","level":2,"score":0.37450000643730164},{"id":"https://openalex.org/C22789450","wikidata":"https://www.wikidata.org/wiki/Q420904","display_name":"Singular value decomposition","level":2,"score":0.3693999946117401},{"id":"https://openalex.org/C82876162","wikidata":"https://www.wikidata.org/wiki/Q17096504","display_name":"Latency (audio)","level":2,"score":0.367900013923645},{"id":"https://openalex.org/C124681953","wikidata":"https://www.wikidata.org/wiki/Q339062","display_name":"Decomposition","level":2,"score":0.35589998960494995},{"id":"https://openalex.org/C57493831","wikidata":"https://www.wikidata.org/wiki/Q3134666","display_name":"Projection (relational algebra)","level":2,"score":0.3447999954223633},{"id":"https://openalex.org/C186370098","wikidata":"https://www.wikidata.org/wiki/Q442787","display_name":"Energy (signal processing)","level":2,"score":0.3325999975204468},{"id":"https://openalex.org/C19768560","wikidata":"https://www.wikidata.org/wiki/Q320727","display_name":"Dependency (UML)","level":2,"score":0.3271999955177307},{"id":"https://openalex.org/C50644808","wikidata":"https://www.wikidata.org/wiki/Q192776","display_name":"Artificial neural network","level":2,"score":0.32679998874664307},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.3102000057697296},{"id":"https://openalex.org/C96324660","wikidata":"https://www.wikidata.org/wiki/Q205446","display_name":"Dataflow","level":2,"score":0.3010999858379364},{"id":"https://openalex.org/C94835093","wikidata":"https://www.wikidata.org/wiki/Q3113333","display_name":"Data compression ratio","level":5,"score":0.2994000017642975},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.28949999809265137},{"id":"https://openalex.org/C459310","wikidata":"https://www.wikidata.org/wiki/Q117801","display_name":"Computational science","level":1,"score":0.27880001068115234},{"id":"https://openalex.org/C137293760","wikidata":"https://www.wikidata.org/wiki/Q3621696","display_name":"Language model","level":2,"score":0.2734000086784363},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.2734000086784363},{"id":"https://openalex.org/C81845259","wikidata":"https://www.wikidata.org/wiki/Q290117","display_name":"Quadratic programming","level":2,"score":0.2685000002384186},{"id":"https://openalex.org/C42704193","wikidata":"https://www.wikidata.org/wiki/Q7851097","display_name":"Tucker decomposition","level":4,"score":0.2646999955177307},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.2644999921321869},{"id":"https://openalex.org/C13164978","wikidata":"https://www.wikidata.org/wiki/Q600158","display_name":"Hardware acceleration","level":3,"score":0.26350000500679016},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.25279998779296875},{"id":"https://openalex.org/C125411270","wikidata":"https://www.wikidata.org/wiki/Q18653","display_name":"Encoding (memory)","level":2,"score":0.25270000100135803},{"id":"https://openalex.org/C80444323","wikidata":"https://www.wikidata.org/wiki/Q2878974","display_name":"Theoretical computer science","level":1,"score":0.25200000405311584},{"id":"https://openalex.org/C2779530757","wikidata":"https://www.wikidata.org/wiki/Q1207505","display_name":"Quality (philosophy)","level":2,"score":0.2515000104904175}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3801487.3801814","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3801487.3801814","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 23rd ACM International Conference on Computing Frontiers","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3801487.3801814","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3801487.3801814","pdf_url":null,"source":null,"license":"cc-by","license_id":"https://openalex.org/licenses/cc-by","version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"Proceedings of the 23rd ACM International Conference on Computing Frontiers","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":13,"referenced_works":["https://openalex.org/W2168113051","https://openalex.org/W2294097286","https://openalex.org/W2798878556","https://openalex.org/W3027968530","https://openalex.org/W3159727696","https://openalex.org/W3196923642","https://openalex.org/W4312569226","https://openalex.org/W4389524555","https://openalex.org/W4406164090","https://openalex.org/W4415593610","https://openalex.org/W4417530628","https://openalex.org/W7106151227","https://openalex.org/W7133227460"],"related_works":[],"abstract_inverted_index":{"Neuromorphic":[0],"large":[1],"language":[2,223],"models":[3,224],"replace":[4],"quadratic":[5,43],"self-attention":[6],"with":[7,105,188],"stateful":[8],"recurrence":[9,81],"to":[10,37,109,169],"achieve":[11],"linear-time":[12],"inference,":[13],"but":[14],"the":[15,77,111,150,170,185],"resulting":[16],"spiking":[17,67],"Receptance":[18],"Weighted":[19],"Key":[20],"Value":[21],"(spiking":[22],"RWKV,":[23],"SRWKV)":[24],"attention":[25,84,102],"kernel":[26,187],"introduces":[27],"long,":[28],"numerically":[29,78],"sensitive":[30],"dependency":[31],"chains":[32],"that":[33,52,135,213],"are":[34],"poorly":[35],"matched":[36],"general-purpose":[38],"processors.":[39],"Although":[40],"SRWKV":[41,83,101,186],"reduces":[42],"complexity,":[44],"its":[45],"recurrent":[46,114],"structure":[47],"imposes":[48],"strict":[49],"ordering":[50],"constraints":[51],"limit":[53],"parallelism":[54],"and":[55,73,103,131,143,152,164,177,199],"make":[56],"execution":[57,99],"latency-dominated":[58],"in":[59,85],"practice.":[60],"Prior":[61],"neuromorphic":[62,222],"accelerators":[63],"primarily":[64],"target":[65],"generic":[66],"workloads":[68],"or":[69],"domain-agnostic":[70],"neural":[71],"primitives,":[72],"do":[74],"not":[75],"address":[76],"continuous,":[79],"latency-critical":[80],"underlying":[82],"SpikeGPT.":[86],"This":[87],"article":[88],"proposes":[89],"SABRE,":[90],"a":[91,136,189,200],"compression-aware":[92],"BF16":[93],"accelerator":[94],"designed":[95],"specifically":[96],"for":[97,149,219],"efficient":[98],"of":[100,113,118,192,203],"co-designed":[104],"activation-aware":[106,119],"low-rank":[107],"compression":[108,125,141,154],"reduce":[110],"cost":[112],"projection":[115],"operations.":[116],"Evaluation":[117],"singular":[120],"value":[121],"decomposition":[122],"at":[123,157,180,226],"three":[124],"operating":[126,181],"points":[127],"is":[128,133,147,217],"carried":[129],"out,":[130],"it":[132],"observed":[134],"clear":[137],"trade-off":[138],"exists":[139],"between":[140],"efficiency":[142],"model":[144,160],"quality.":[145],"SABRE":[146,183],"implemented":[148],"30%":[151],"50%":[153],"regimes":[155],"and,":[156],"moderate":[158],"compression,":[159],"quality":[161],"on":[162],"WikiText-2":[163],"SST-2":[165],"degrades":[166],"modestly":[167],"relative":[168],"uncompressed":[171],"baseline.":[172],"Using":[173],"cycle-accurate":[174],"RTL":[175],"modeling":[176],"post-place-and-route":[178],"analysis":[179],"frequency,":[182],"executes":[184],"per-layer":[190],"latency":[191],"2.44":[193],"ms":[194],"(410":[195],"tokens/s":[196],"per":[197,206,208],"layer)":[198],"dynamic":[201],"energy":[202],"4.60":[204],"mJ":[205],"token":[207],"layer.":[209],"These":[210],"results":[211],"demonstrate":[212],"workload-specific":[214],"hardware\u2013compression":[215],"co-design":[216],"essential":[218],"making":[220],"SRWKV-based":[221],"practical":[225],"realistic":[227],"scales.":[228]},"counts_by_year":[],"updated_date":"2026-06-28T06:20:29.262167","created_date":"2026-06-28T00:00:00"}