{"id":"https://openalex.org/W7164171983","doi":"https://doi.org/10.1109/fccm68464.2026.00034","title":"TransDot: An Area-efficient Reconfigurable Floating-Point Unit for Trans-Precision Dot-Product Accumulation for FPGA AI Engines","display_name":"TransDot: An Area-efficient Reconfigurable Floating-Point Unit for Trans-Precision Dot-Product Accumulation for FPGA AI Engines","publication_year":2026,"publication_date":"2026-05-13","ids":{"openalex":"https://openalex.org/W7164171983","doi":"https://doi.org/10.1109/fccm68464.2026.00034"},"language":null,"primary_location":{"id":"doi:10.1109/fccm68464.2026.00034","is_oa":false,"landing_page_url":"https://doi.org/10.1109/fccm68464.2026.00034","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE 34th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","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/A5051762904","display_name":"J H Wang","orcid":null},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiayi Wang","raw_affiliation_strings":["University of Washington"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5026647054","display_name":"Maohua Nie","orcid":"https://orcid.org/0000-0003-3515-7764"},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Maohua Nie","raw_affiliation_strings":["University of Washington"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5101375769","display_name":"Sin-Chen Lin","orcid":null},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sin-Chen Lin","raw_affiliation_strings":["University of Washington"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5138298047","display_name":"C.-J. Richard Shi","orcid":null},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"C.-J. Richard Shi","raw_affiliation_strings":["University of Washington"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington","institution_ids":["https://openalex.org/I201448701"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5138350638","display_name":"Ang Li","orcid":null},"institutions":[{"id":"https://openalex.org/I201448701","display_name":"University of Washington","ror":"https://ror.org/00cvxb145","country_code":"US","type":"education","lineage":["https://openalex.org/I201448701"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Ang Li","raw_affiliation_strings":["University of Washington"],"raw_orcid":null,"affiliations":[{"raw_affiliation_string":"University of Washington","institution_ids":["https://openalex.org/I201448701"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I201448701"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.85891262,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":null,"issue":null,"first_page":"171","last_page":"175"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11697","display_name":"Numerical Methods and Algorithms","score":0.84579998254776,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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/T11697","display_name":"Numerical Methods and Algorithms","score":0.84579998254776,"subfield":{"id":"https://openalex.org/subfields/1703","display_name":"Computational Theory and Mathematics"},"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":0.053199999034404755,"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/T11522","display_name":"VLSI and FPGA Design Techniques","score":0.013000000268220901,"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/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.41679999232292175},{"id":"https://openalex.org/keywords/unit","display_name":"Unit (ring theory)","score":0.4133000075817108},{"id":"https://openalex.org/keywords/key","display_name":"Key (lock)","score":0.2874999940395355},{"id":"https://openalex.org/keywords/reconfigurable-computing","display_name":"Reconfigurable computing","score":0.25609999895095825}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.5464000105857849},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.41839998960494995},{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.41679999232292175},{"id":"https://openalex.org/C122637931","wikidata":"https://www.wikidata.org/wiki/Q118084","display_name":"Unit (ring theory)","level":2,"score":0.4133000075817108},{"id":"https://openalex.org/C9390403","wikidata":"https://www.wikidata.org/wiki/Q3966","display_name":"Computer hardware","level":1,"score":0.3418000042438507},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.30660000443458557},{"id":"https://openalex.org/C26517878","wikidata":"https://www.wikidata.org/wiki/Q228039","display_name":"Key (lock)","level":2,"score":0.2874999940395355},{"id":"https://openalex.org/C142962650","wikidata":"https://www.wikidata.org/wiki/Q240838","display_name":"Reconfigurable computing","level":3,"score":0.25609999895095825},{"id":"https://openalex.org/C99498987","wikidata":"https://www.wikidata.org/wiki/Q2210247","display_name":"Noise (video)","level":3,"score":0.2484000027179718},{"id":"https://openalex.org/C79403827","wikidata":"https://www.wikidata.org/wiki/Q3988","display_name":"Real-time computing","level":1,"score":0.24799999594688416}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/fccm68464.2026.00034","is_oa":false,"landing_page_url":"https://doi.org/10.1109/fccm68464.2026.00034","pdf_url":null,"source":null,"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"2026 IEEE 34th Annual International Symposium on Field-Programmable Custom Computing Machines (FCCM)","raw_type":"proceedings-article"}],"best_oa_location":null,"sustainable_development_goals":[{"display_name":"Affordable and clean energy","score":0.49336233735084534,"id":"https://metadata.un.org/sdg/7"}],"awards":[],"funders":[],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":10,"referenced_works":["https://openalex.org/W2008424999","https://openalex.org/W2026445983","https://openalex.org/W2313356641","https://openalex.org/W2606722458","https://openalex.org/W2914973130","https://openalex.org/W3036878841","https://openalex.org/W3113606433","https://openalex.org/W4391944268","https://openalex.org/W4404954859","https://openalex.org/W4411486388"],"related_works":[],"abstract_inverted_index":{"Commercial":[0],"FPGAs,":[1],"such":[2,88],"as":[3,89],"AMD":[4,226],"Versal":[5,227],"devices,":[6],"increasingly":[7],"incorporate":[8],"AI":[9,228],"engines":[10],"that":[11,127,162,217],"exploit":[12],"low-precision":[13,100],"packed-SIMD":[14],"fused":[15],"multiply\u2013accumulate":[16],"(FMA)":[17],"to":[18,35,71,210],"achieve":[19],"proportional":[20],"throughput":[21,172],"gains.":[22],"However,":[23],"trans-precision":[24,133],"FMA":[25,98,105,131],"(e.g.,":[26,58],"multiplying":[27],"two":[28,63],"FP16":[29,183],"numbers":[30],"and":[31,67,81,95,114,132,150,169,178,185,201],"adding":[32,68],"their":[33],"result":[34,70],"an":[36,72,202],"FP32":[37,73,156,176],"accumulator),":[38],"which":[39,107],"preserves":[40],"numerical":[41],"stability":[42],"by":[43,50,102],"accumulating":[44],"in":[45,182,189,206,224],"higher":[46],"precision,":[47],"remains":[48],"bottlenecked":[49],"the":[51,78,116,142,193,211],"highest-precision,":[52],"lowest-throughput":[53],"operation.":[54],"Dot-product":[55],"accumulation":[56,154],"(DPA)":[57],"performing":[59],"a":[60,124,136],"dot-product":[61,153,207],"on":[62,99,199],"4-element":[64],"FP8":[65,190],"vectors":[66],"its":[69],"accumulator)":[74],"can":[75],"fully":[76],"utilize":[77],"input/output":[79],"bandwidth":[80],"computational":[82],"resources.":[83],"Existing":[84],"flexible":[85],"open-source":[86],"FPUs,":[87],"FPnew,":[90],"do":[91],"not":[92],"support":[93],"DPA":[94,119,134,174,184],"implement":[96],"SIMD":[97,130],"formats":[101],"replicating":[103],"independent":[104],"lanes,":[106],"increases":[108],"area,":[109],"underutilizes":[110],"shared":[111],"arithmetic":[112],"resources,":[113],"complicates":[115],"integration":[117],"of":[118,195],"operations.This":[120],"paper":[121],"presents":[122],"TransDot,":[123],"reconfigurable":[125,138,158],"FPU":[126],"unifies":[128],"multi-precision":[129],"within":[135],"shared,":[137],"datapath.":[139],"TransDot":[140,163],"extends":[141],"baseline":[143],"design":[144,220],"with":[145,175],"2-term":[146],"FP16,":[147,166],"4-term":[148],"FP8,":[149,168],"8-term":[151],"FP4":[152,171],"into":[155],"using":[157],"subcomponents.":[159],"Evaluation":[160],"shows":[161],"delivers":[164],"2\u00d7":[165],"4\u00d7":[167],"8\u00d7":[170],"via":[173],"accumulation,":[177],"1.46\u00d7":[179],"area":[180,187,198],"efficiency":[181,188],"2.92\u00d7":[186],"DPA,":[191],"at":[192],"cost":[194],"37.3%":[196],"larger":[197],"average":[200],"additional":[203],"pipeline":[204],"stage":[205],"mode":[208],"compared":[209],"FPnew":[212],"baseline.":[213],"These":[214],"results":[215],"demonstrate":[216],"TransDot\u2019s":[218],"area-efficient":[219],"enables":[221],"scalable":[222],"deployment":[223],"next-generation":[225],"engines.":[229]},"counts_by_year":[],"updated_date":"2026-06-26T08:34:08.712188","created_date":"2026-06-11T00:00:00"}