{"id":"https://openalex.org/W4414197462","doi":"https://doi.org/10.1109/dac63849.2025.11132816","title":"DuoQ: A DSP Utilization-aware and Outlier-free Quantization for FPGA-based LLMs Acceleration","display_name":"DuoQ: A DSP Utilization-aware and Outlier-free Quantization for FPGA-based LLMs Acceleration","publication_year":2025,"publication_date":"2025-06-22","ids":{"openalex":"https://openalex.org/W4414197462","doi":"https://doi.org/10.1109/dac63849.2025.11132816"},"language":"en","primary_location":{"id":"doi:10.1109/dac63849.2025.11132816","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11132816","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (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/A5100311074","display_name":"Zhuoquan Yu","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Zhuoquan Yu","raw_affiliation_strings":["Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"],"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5028503613","display_name":"Huidong Ji","orcid":"https://orcid.org/0009-0005-7202-1681"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Huidong Ji","raw_affiliation_strings":["Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"],"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039207695","display_name":"Yue Cao","orcid":"https://orcid.org/0009-0001-4050-6415"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yue Cao","raw_affiliation_strings":["Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"],"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5120391756","display_name":"Junfu Wu","orcid":"https://orcid.org/0000-0002-3394-1507"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Junfu Wu","raw_affiliation_strings":["Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"],"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5111296369","display_name":"Xiaoze Yan","orcid":null},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Xiaoze Yan","raw_affiliation_strings":["Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"],"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5024591419","display_name":"Lirong Zheng","orcid":"https://orcid.org/0000-0003-0888-6769"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Lirong Zheng","raw_affiliation_strings":["Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"],"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5043039025","display_name":"Zhuo Zou","orcid":"https://orcid.org/0000-0002-8546-1329"},"institutions":[{"id":"https://openalex.org/I24943067","display_name":"Fudan University","ror":"https://ror.org/013q1eq08","country_code":"CN","type":"education","lineage":["https://openalex.org/I24943067"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhuo Zou","raw_affiliation_strings":["Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"],"affiliations":[{"raw_affiliation_string":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China","institution_ids":["https://openalex.org/I24943067"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":7,"corresponding_author_ids":["https://openalex.org/A5100311074"],"corresponding_institution_ids":["https://openalex.org/I24943067"],"apc_list":null,"apc_paid":null,"fwci":2.1983,"has_fulltext":false,"cited_by_count":1,"citation_normalized_percentile":{"value":0.87910725,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":96,"max":98},"biblio":{"volume":null,"issue":null,"first_page":"1","last_page":"7"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11032","display_name":"VLSI and Analog Circuit Testing","score":0.9926000237464905,"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/T11032","display_name":"VLSI and Analog Circuit Testing","score":0.9926000237464905,"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/T10904","display_name":"Embedded Systems Design Techniques","score":0.9714000225067139,"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/T11338","display_name":"Advancements in Photolithography Techniques","score":0.9627000093460083,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/quantization","display_name":"Quantization (signal processing)","score":0.8838000297546387},{"id":"https://openalex.org/keywords/security-token","display_name":"Security token","score":0.5834000110626221},{"id":"https://openalex.org/keywords/encoder","display_name":"Encoder","score":0.5630999803543091},{"id":"https://openalex.org/keywords/outlier","display_name":"Outlier","score":0.5270000100135803},{"id":"https://openalex.org/keywords/digital-signal-processing","display_name":"Digital signal processing","score":0.4523000121116638},{"id":"https://openalex.org/keywords/software","display_name":"Software","score":0.3921000063419342},{"id":"https://openalex.org/keywords/nonlinear-system","display_name":"Nonlinear system","score":0.38760000467300415},{"id":"https://openalex.org/keywords/linde\u2013buzo\u2013gray-algorithm","display_name":"Linde\u2013Buzo\u2013Gray algorithm","score":0.382999986410141}],"concepts":[{"id":"https://openalex.org/C28855332","wikidata":"https://www.wikidata.org/wiki/Q198099","display_name":"Quantization (signal processing)","level":2,"score":0.8838000297546387},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.6198999881744385},{"id":"https://openalex.org/C48145219","wikidata":"https://www.wikidata.org/wiki/Q1335365","display_name":"Security token","level":2,"score":0.5834000110626221},{"id":"https://openalex.org/C118505674","wikidata":"https://www.wikidata.org/wiki/Q42586063","display_name":"Encoder","level":2,"score":0.5630999803543091},{"id":"https://openalex.org/C79337645","wikidata":"https://www.wikidata.org/wiki/Q779824","display_name":"Outlier","level":2,"score":0.5270000100135803},{"id":"https://openalex.org/C84462506","wikidata":"https://www.wikidata.org/wiki/Q173142","display_name":"Digital signal processing","level":2,"score":0.4523000121116638},{"id":"https://openalex.org/C11413529","wikidata":"https://www.wikidata.org/wiki/Q8366","display_name":"Algorithm","level":1,"score":0.446399986743927},{"id":"https://openalex.org/C113775141","wikidata":"https://www.wikidata.org/wiki/Q428691","display_name":"Computer engineering","level":1,"score":0.4207000136375427},{"id":"https://openalex.org/C2777904410","wikidata":"https://www.wikidata.org/wiki/Q7397","display_name":"Software","level":2,"score":0.3921000063419342},{"id":"https://openalex.org/C158622935","wikidata":"https://www.wikidata.org/wiki/Q660848","display_name":"Nonlinear system","level":2,"score":0.38760000467300415},{"id":"https://openalex.org/C93372532","wikidata":"https://www.wikidata.org/wiki/Q6552455","display_name":"Linde\u2013Buzo\u2013Gray algorithm","level":3,"score":0.382999986410141},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3720000088214874},{"id":"https://openalex.org/C105339364","wikidata":"https://www.wikidata.org/wiki/Q2297740","display_name":"Software deployment","level":2,"score":0.3653999865055084},{"id":"https://openalex.org/C77618280","wikidata":"https://www.wikidata.org/wiki/Q1155772","display_name":"Scheme (mathematics)","level":2,"score":0.35600000619888306},{"id":"https://openalex.org/C2781089630","wikidata":"https://www.wikidata.org/wiki/Q21856745","display_name":"Realization (probability)","level":2,"score":0.352400004863739},{"id":"https://openalex.org/C2742236","wikidata":"https://www.wikidata.org/wiki/Q924713","display_name":"Efficient energy use","level":2,"score":0.33169999718666077},{"id":"https://openalex.org/C58870171","wikidata":"https://www.wikidata.org/wiki/Q116076","display_name":"CORDIC","level":3,"score":0.31450000405311584},{"id":"https://openalex.org/C47446073","wikidata":"https://www.wikidata.org/wiki/Q5165890","display_name":"Control theory (sociology)","level":3,"score":0.31290000677108765},{"id":"https://openalex.org/C24326235","wikidata":"https://www.wikidata.org/wiki/Q126095","display_name":"Electronic engineering","level":1,"score":0.3118000030517578},{"id":"https://openalex.org/C173018170","wikidata":"https://www.wikidata.org/wiki/Q165678","display_name":"Microcontroller","level":2,"score":0.30869999527931213},{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.2770000100135803},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.2619999945163727},{"id":"https://openalex.org/C78548338","wikidata":"https://www.wikidata.org/wiki/Q2493","display_name":"Data compression","level":2,"score":0.25949999690055847},{"id":"https://openalex.org/C59883199","wikidata":"https://www.wikidata.org/wiki/Q1826438","display_name":"Linear predictive coding","level":3,"score":0.25360000133514404}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/dac63849.2025.11132816","is_oa":false,"landing_page_url":"https://doi.org/10.1109/dac63849.2025.11132816","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2025 62nd ACM/IEEE Design Automation Conference (DAC)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[],"funders":[{"id":"https://openalex.org/F4320335777","display_name":"National Key Research and Development Program of China","ror":null}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W2613938013","https://openalex.org/W2762374354","https://openalex.org/W2905101520","https://openalex.org/W2946609015","https://openalex.org/W2998617917","https://openalex.org/W3017024317","https://openalex.org/W3100985894","https://openalex.org/W3126267573","https://openalex.org/W3171007011","https://openalex.org/W3176468986","https://openalex.org/W3189877953","https://openalex.org/W3194676777","https://openalex.org/W4293025109","https://openalex.org/W4308083513","https://openalex.org/W4321637080","https://openalex.org/W4366341968","https://openalex.org/W4390871466","https://openalex.org/W4390946831","https://openalex.org/W4393178455","https://openalex.org/W4393578753","https://openalex.org/W4393949386","https://openalex.org/W4402727233","https://openalex.org/W4403278953","https://openalex.org/W4403864152","https://openalex.org/W4404133703"],"related_works":[],"abstract_inverted_index":{"Quantization":[0],"enables":[1],"efficient":[2],"deployment":[3],"of":[4,14,19,104,111],"large":[5],"language":[6],"models":[7],"(LLMs)":[8],"on":[9],"FPGAs,":[10],"but":[11],"its":[12],"presence":[13],"outliers":[15,28,57],"affects":[16],"the":[17,20,67,79,109],"accuracy":[18,149],"quantized":[21],"model.":[22],"Existing":[23],"methods":[24],"mainly":[25],"deal":[26],"with":[27,74,123,125],"through":[29,58,90],"channel-wise":[30],"or":[31],"tokenwise":[32],"isolation":[33],"and":[34,62,82,97,114,131,141,156],"encoding,":[35],"which":[36],"leads":[37],"to":[38,138,152],"expensive":[39],"dynamic":[40,115],"quantization.":[41],"To":[42],"address":[43],"this":[44],"problem,":[45],"we":[46],"introduce":[47],"DuoQ,":[48],"an":[49],"FPGA-oriented":[50],"algorithm-hardware":[51],"co-design":[52,78],"framework.":[53],"DuoQ":[54,86,146],"effectively":[55],"eliminates":[56],"learnable":[59],"equivalent":[60],"transformations":[61],"low-semantic":[63],"token":[64,116],"awareness":[65],"in":[66,108],"quantization":[68,73,80],"scheme":[69],"part,":[70],"facilitating":[71],"per-tensor":[72],"4":[75],"-bits.":[76],"We":[77],"algorithm":[81],"hardware":[83,157],"architecture.":[84],"Specifically,":[85],"accelerates":[87],"end-to-end":[88],"LLM":[89],"a":[91],"novel":[92],"DSP-aware":[93],"PE":[94],"unit":[95],"design":[96],"encoder":[98],"design.":[99],"In":[100,144],"addition,":[101,145],"two":[102],"types":[103],"post-processing":[105],"units":[106],"assist":[107],"realization":[110],"nonlinear":[112],"functions":[113],"awareness.":[117],"Experimental":[118],"results":[119],"show":[120],"that":[121],"compared":[122,151],"platforms":[124],"different":[126],"architectures,":[127],"DuoQ\u2019s":[128],"computational":[129],"efficiency":[130,133],"energy":[132],"are":[134],"improved":[135],"by":[136],"up":[137],"$8.8":[139],"\\times$":[140],"$23.45":[142],"\\times$.":[143],"has":[147],"achieved":[148],"improvements":[150],"other":[153],"outlieraware":[154],"software":[155],"works.":[158]},"counts_by_year":[{"year":2026,"cited_by_count":1}],"updated_date":"2026-04-09T08:11:56.329763","created_date":"2025-10-10T00:00:00"}