{"id":"https://openalex.org/W4411268230","doi":"https://doi.org/10.1145/3735452.3735523","title":"SPARQ: An Accelerator Architecture for Large Language Models with Joint Sparsity and Quantization Techniques","display_name":"SPARQ: An Accelerator Architecture for Large Language Models with Joint Sparsity and Quantization Techniques","publication_year":2025,"publication_date":"2025-06-13","ids":{"openalex":"https://openalex.org/W4411268230","doi":"https://doi.org/10.1145/3735452.3735523"},"language":"en","primary_location":{"id":"doi:10.1145/3735452.3735523","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3735452.3735523","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3735452.3735523","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 26th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems","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/10.1145/3735452.3735523","any_repository_has_fulltext":null},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Seonggyu Choi","orcid":"https://orcid.org/0009-0003-3191-610X"},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Seonggyu Choi","raw_affiliation_strings":["Sungkyunkwan University, Suwon, Republic of Korea"],"raw_orcid":"https://orcid.org/0009-0003-3191-610X","affiliations":[{"raw_affiliation_string":"Sungkyunkwan University, Suwon, Republic of Korea","institution_ids":["https://openalex.org/I848706"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5061945152","display_name":"Hyungmin Cho","orcid":"https://orcid.org/0000-0001-8705-7066"},"institutions":[{"id":"https://openalex.org/I848706","display_name":"Sungkyunkwan University","ror":"https://ror.org/04q78tk20","country_code":"KR","type":"education","lineage":["https://openalex.org/I848706"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Hyungmin Cho","raw_affiliation_strings":["Sungkyunkwan University, Suwon, Republic of Korea"],"raw_orcid":"https://orcid.org/0000-0001-8705-7066","affiliations":[{"raw_affiliation_string":"Sungkyunkwan University, Suwon, Republic of Korea","institution_ids":["https://openalex.org/I848706"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I848706"],"apc_list":null,"apc_paid":null,"fwci":1.2409,"has_fulltext":true,"cited_by_count":1,"citation_normalized_percentile":{"value":0.72768833,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":94,"max":97},"biblio":{"volume":null,"issue":null,"first_page":"3","last_page":"15"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9962000250816345,"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":0.9962000250816345,"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/T10028","display_name":"Topic Modeling","score":0.9904999732971191,"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/T10201","display_name":"Speech Recognition and Synthesis","score":0.9902999997138977,"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/computer-science","display_name":"Computer science","score":0.7013354301452637},{"id":"https://openalex.org/keywords/architecture","display_name":"Architecture","score":0.6419203281402588},{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.5765160918235779},{"id":"https://openalex.org/keywords/joint","display_name":"Joint (building)","score":0.5457451343536377},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.3435165584087372},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.34092819690704346},{"id":"https://openalex.org/keywords/algorithm","display_name":"Algorithm","score":0.16762423515319824},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.1137588620185852},{"id":"https://openalex.org/keywords/history","display_name":"History","score":0.09716126322746277},{"id":"https://openalex.org/keywords/structural-engineering","display_name":"Structural engineering","score":0.05952683091163635}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.7013354301452637},{"id":"https://openalex.org/C123657996","wikidata":"https://www.wikidata.org/wiki/Q12271","display_name":"Architecture","level":2,"score":0.6419203281402588},{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.5765160918235779},{"id":"https://openalex.org/C18555067","wikidata":"https://www.wikidata.org/wiki/Q8375051","display_name":"Joint (building)","level":2,"score":0.5457451343536377},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.3435165584087372},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.34092819690704346},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.16762423515319824},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.1137588620185852},{"id":"https://openalex.org/C95457728","wikidata":"https://www.wikidata.org/wiki/Q309","display_name":"History","level":0,"score":0.09716126322746277},{"id":"https://openalex.org/C66938386","wikidata":"https://www.wikidata.org/wiki/Q633538","display_name":"Structural engineering","level":1,"score":0.05952683091163635},{"id":"https://openalex.org/C166957645","wikidata":"https://www.wikidata.org/wiki/Q23498","display_name":"Archaeology","level":1,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3735452.3735523","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3735452.3735523","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3735452.3735523","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 26th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems","raw_type":"proceedings-article"}],"best_oa_location":{"id":"doi:10.1145/3735452.3735523","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3735452.3735523","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3735452.3735523","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 26th ACM SIGPLAN/SIGBED International Conference on Languages, Compilers, and Tools for Embedded Systems","raw_type":"proceedings-article"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G3184925354","display_name":null,"funder_award_id":"RS-2024-00398745","funder_id":"https://openalex.org/F4320335489","funder_display_name":"Institute for Information and Communications Technology Promotion"},{"id":"https://openalex.org/G3812690899","display_name":null,"funder_award_id":"RS-2023","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G6744989115","display_name":null,"funder_award_id":"RS-2024-00398745","funder_id":"https://openalex.org/F4320328359","funder_display_name":"Ministry of Science and ICT, South Korea"},{"id":"https://openalex.org/G7393106735","display_name":null,"funder_award_id":"RS-2023-00251287","funder_id":"https://openalex.org/F4320322120","funder_display_name":"National Research Foundation of Korea"}],"funders":[{"id":"https://openalex.org/F4320320671","display_name":"National Research Foundation","ror":"https://ror.org/05s0g1g46"},{"id":"https://openalex.org/F4320322120","display_name":"National Research Foundation of Korea","ror":"https://ror.org/013aysd81"},{"id":"https://openalex.org/F4320328359","display_name":"Ministry of Science and ICT, South Korea","ror":"https://ror.org/01wpjm123"},{"id":"https://openalex.org/F4320335489","display_name":"Institute for Information and Communications Technology Promotion","ror":"https://ror.org/01g0hqq23"}],"has_content":{"pdf":true,"grobid_xml":true},"content_urls":{"pdf":"https://content.openalex.org/works/W4411268230.pdf","grobid_xml":"https://content.openalex.org/works/W4411268230.grobid-xml"},"referenced_works_count":11,"referenced_works":["https://openalex.org/W2560674852","https://openalex.org/W2725159389","https://openalex.org/W2979691890","https://openalex.org/W3012178976","https://openalex.org/W3185702163","https://openalex.org/W4245659846","https://openalex.org/W4291653336","https://openalex.org/W4390097933","https://openalex.org/W4390098256","https://openalex.org/W4404133735","https://openalex.org/W6930885647"],"related_works":["https://openalex.org/W1996130883","https://openalex.org/W2748574964","https://openalex.org/W2888483922","https://openalex.org/W4396737233","https://openalex.org/W2588198209","https://openalex.org/W2367747139","https://openalex.org/W1909006023","https://openalex.org/W4205824991","https://openalex.org/W3200723557","https://openalex.org/W2038503502"],"abstract_inverted_index":{"Large":[0],"Language":[1],"Models":[2],"(LLMs)":[3],"have":[4],"demonstrated":[5,84],"unprecedented":[6],"capabilities":[7],"in":[8,85],"text":[9],"generation,":[10],"translation,":[11],"and":[12,29,47,61,74,90,105],"summarization":[13],"tasks.":[14],"However,":[15],"their":[16,25],"deployment":[17],"on":[18,80,88],"resource-constrained":[19],"systems":[20],"remains":[21],"challenging":[22],"due":[23],"to":[24,49,99,112],"large":[26],"parameter":[27],"sizes":[28],"high":[30],"computational":[31],"demands.":[32],"To":[33],"address":[34],"this,":[35],"we":[36],"propose":[37],"SPARQ,":[38],"a":[39,62],"specialized":[40],"accelerator":[41],"architecture":[42,65],"that":[43,95],"leverages":[44],"both":[45],"sparsity":[46],"quantization":[48],"optimize":[50],"LLM":[51],"inference.":[52],"By":[53],"integrating":[54],"multiply-accumulate":[55],"units":[56],"tailored":[57],"for":[58,116],"quantized":[59],"operations":[60],"systolic":[63],"array":[64],"supporting":[66],"N:M":[67],"semi-structured":[68],"sparsity,":[69],"SPARQ":[70,96],"significantly":[71],"enhances":[72],"area":[73,103],"energy":[75,109],"efficiency":[76,104,110],"with":[77],"minimal":[78],"impact":[79],"model":[81],"quality,":[82],"as":[83],"prior":[86],"work":[87],"GPTQ":[89],"SparseGPT.":[91],"Our":[92],"evaluations":[93],"demonstrate":[94],"achieves":[97],"up":[98],"1.53":[100],"times":[101,107],"greater":[102],"1.58":[106],"better":[108],"compared":[111],"the":[113],"baseline,":[114],"particularly":[115],"larger":[117],"models.":[118]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-08-21T09:56:20.448147","created_date":"2025-10-10T00:00:00"}