{"id":"https://openalex.org/W4410374386","doi":"https://doi.org/10.1145/3728470","title":"Diff-Acc: An Efficient FPGA Accelerator for Unconditional Diffusion Models","display_name":"Diff-Acc: An Efficient FPGA Accelerator for Unconditional Diffusion Models","publication_year":2025,"publication_date":"2025-05-14","ids":{"openalex":"https://openalex.org/W4410374386","doi":"https://doi.org/10.1145/3728470"},"language":"en","primary_location":{"id":"doi:10.1145/3728470","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3728470","pdf_url":null,"source":{"id":"https://openalex.org/S136160450","display_name":"ACM Transactions on Embedded Computing Systems","issn_l":"1539-9087","issn":["1539-9087","1558-3465"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Embedded Computing Systems","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/A5087976552","display_name":"Shidi Tang","orcid":"https://orcid.org/0000-0001-5493-7411"},"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":"Shidi Tang","raw_affiliation_strings":["Southeast University","Southeast University,  Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0001-5493-7411","affiliations":[{"raw_affiliation_string":"Southeast University","institution_ids":[]},{"raw_affiliation_string":"Southeast University,  Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100761386","display_name":"Ruiqi Chen","orcid":"https://orcid.org/0000-0001-6837-5675"},"institutions":[{"id":"https://openalex.org/I13469542","display_name":"Vrije Universiteit Brussel","ror":"https://ror.org/006e5kg04","country_code":"BE","type":"education","lineage":["https://openalex.org/I13469542"]}],"countries":["BE"],"is_corresponding":false,"raw_author_name":"Ruiqi Chen","raw_affiliation_strings":["Departement of Electronics and Informatics, Vrije Universiteit Brussel","Departement of Electronics and Informatics, Vrije Universiteit Brussel, Brussels, Belgium"],"raw_orcid":"https://orcid.org/0000-0001-6837-5675","affiliations":[{"raw_affiliation_string":"Departement of Electronics and Informatics, Vrije Universiteit Brussel","institution_ids":["https://openalex.org/I13469542"]},{"raw_affiliation_string":"Departement of Electronics and Informatics, Vrije Universiteit Brussel, Brussels, Belgium","institution_ids":["https://openalex.org/I13469542"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5112616411","display_name":"Rui Liu","orcid":"https://orcid.org/0009-0007-4211-403X"},"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":"Rui Liu","raw_affiliation_strings":["Southeast University","Southeast University,  Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0007-4211-403X","affiliations":[{"raw_affiliation_string":"Southeast University","institution_ids":[]},{"raw_affiliation_string":"Southeast University,  Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5108874948","display_name":"Yuxuan Lv","orcid":"https://orcid.org/0009-0002-9128-1479"},"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":"Yuxuan Lv","raw_affiliation_strings":["Southeast University","Southeast University,  Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0002-9128-1479","affiliations":[{"raw_affiliation_string":"Southeast University","institution_ids":[]},{"raw_affiliation_string":"Southeast University,  Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5117545672","display_name":"Pengwei Zheng","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":"Pengwei Zheng","raw_affiliation_strings":["Southeast University","Southeast University,  Nanjing, China"],"raw_orcid":"https://orcid.org/0009-0004-4633-6121","affiliations":[{"raw_affiliation_string":"Southeast University","institution_ids":[]},{"raw_affiliation_string":"Southeast University,  Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5100317850","display_name":"He Li","orcid":"https://orcid.org/0000-0002-1540-189X"},"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":"He Li","raw_affiliation_strings":["Southeast University","Southeast University,  Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-1540-189X","affiliations":[{"raw_affiliation_string":"Southeast University","institution_ids":[]},{"raw_affiliation_string":"Southeast University,  Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5078395317","display_name":"Ming Ling","orcid":"https://orcid.org/0000-0002-8866-7189"},"institutions":[{"id":"https://openalex.org/I4210090971","display_name":"Southeast University","ror":"https://ror.org/00cf0ab87","country_code":"BD","type":"education","lineage":["https://openalex.org/I4210090971"]},{"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":["BD","CN"],"is_corresponding":false,"raw_author_name":"Ming Ling","raw_affiliation_strings":["School of Integrated Circuits, Southeast University","School of Integrated Circuits, Southeast University, Nanjing, China"],"raw_orcid":"https://orcid.org/0000-0002-8866-7189","affiliations":[{"raw_affiliation_string":"School of Integrated Circuits, Southeast University","institution_ids":["https://openalex.org/I4210090971"]},{"raw_affiliation_string":"School of Integrated Circuits, Southeast University, Nanjing, China","institution_ids":["https://openalex.org/I76569877"]}]}],"institutions":[],"countries_distinct_count":3,"institutions_distinct_count":3,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":1.6074,"has_fulltext":false,"cited_by_count":2,"citation_normalized_percentile":{"value":0.82425159,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":{"min":97,"max":98},"biblio":{"volume":"25","issue":"3","first_page":"1","last_page":"21"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T11206","display_name":"Model Reduction and Neural Networks","score":0.9914000034332275,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},"topics":[{"id":"https://openalex.org/T11206","display_name":"Model Reduction and Neural Networks","score":0.9914000034332275,"subfield":{"id":"https://openalex.org/subfields/3109","display_name":"Statistical and Nonlinear Physics"},"field":{"id":"https://openalex.org/fields/31","display_name":"Physics and Astronomy"},"domain":{"id":"https://openalex.org/domains/3","display_name":"Physical Sciences"}},{"id":"https://openalex.org/T10363","display_name":"Low-power high-performance VLSI design","score":0.9854000210762024,"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/T11697","display_name":"Numerical Methods and Algorithms","score":0.9785000085830688,"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"}}],"keywords":[{"id":"https://openalex.org/keywords/field-programmable-gate-array","display_name":"Field-programmable gate array","score":0.7079797387123108},{"id":"https://openalex.org/keywords/diffusion","display_name":"Diffusion","score":0.5302590727806091},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.46891292929649353},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.4561833441257477},{"id":"https://openalex.org/keywords/embedded-system","display_name":"Embedded system","score":0.30744433403015137},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.21858417987823486},{"id":"https://openalex.org/keywords/thermodynamics","display_name":"Thermodynamics","score":0.04331636428833008}],"concepts":[{"id":"https://openalex.org/C42935608","wikidata":"https://www.wikidata.org/wiki/Q190411","display_name":"Field-programmable gate array","level":2,"score":0.7079797387123108},{"id":"https://openalex.org/C69357855","wikidata":"https://www.wikidata.org/wiki/Q163214","display_name":"Diffusion","level":2,"score":0.5302590727806091},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.46891292929649353},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.4561833441257477},{"id":"https://openalex.org/C149635348","wikidata":"https://www.wikidata.org/wiki/Q193040","display_name":"Embedded system","level":1,"score":0.30744433403015137},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.21858417987823486},{"id":"https://openalex.org/C97355855","wikidata":"https://www.wikidata.org/wiki/Q11473","display_name":"Thermodynamics","level":1,"score":0.04331636428833008}],"mesh":[],"locations_count":3,"locations":[{"id":"doi:10.1145/3728470","is_oa":false,"landing_page_url":"https://doi.org/10.1145/3728470","pdf_url":null,"source":{"id":"https://openalex.org/S136160450","display_name":"ACM Transactions on Embedded Computing Systems","issn_l":"1539-9087","issn":["1539-9087","1558-3465"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319798","host_organization_name":"Association for Computing Machinery","host_organization_lineage":["https://openalex.org/P4310319798"],"host_organization_lineage_names":["Association for Computing Machinery"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"ACM Transactions on Embedded Computing Systems","raw_type":"journal-article"},{"id":"pmh:oai:vubissmart:VUBISSMART:2000:212869","is_oa":false,"landing_page_url":null,"pdf_url":null,"source":{"id":"https://openalex.org/S4306402573","display_name":"VUBIR (Vrije Universiteit Brussel)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I13469542","host_organization_name":"Vrije Universiteit Brussel","host_organization_lineage":["https://openalex.org/I13469542"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"},{"id":"pmh:oai:vubissmart:VUBISSMART:2000:260012","is_oa":false,"landing_page_url":"https://biblio.vub.ac.be/vubir/diffacc-an-efficient-fpga-accelerator-for-unconditional-diffusion-models(bfce65b9-157f-4f69-8aea-75953a688839).html","pdf_url":null,"source":{"id":"https://openalex.org/S4306402573","display_name":"VUBIR (Vrije Universiteit Brussel)","issn_l":null,"issn":null,"is_oa":false,"is_in_doaj":false,"is_core":false,"host_organization":"https://openalex.org/I13469542","host_organization_name":"Vrije Universiteit Brussel","host_organization_lineage":["https://openalex.org/I13469542"],"host_organization_lineage_names":[],"type":"repository"},"license":null,"license_id":null,"version":"submittedVersion","is_accepted":false,"is_published":false,"raw_source_name":null,"raw_type":"article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G6942830257","display_name":null,"funder_award_id":"BK20230828","funder_id":"https://openalex.org/F4320322769","funder_display_name":"Natural Science Foundation of Jiangsu Province"},{"id":"https://openalex.org/G8713484389","display_name":null,"funder_award_id":"92464301, 62304037","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"}],"funders":[{"id":"https://openalex.org/F4320321001","display_name":"National Natural Science Foundation of China","ror":"https://ror.org/01h0zpd94"},{"id":"https://openalex.org/F4320322769","display_name":"Natural Science Foundation of Jiangsu Province","ror":"https://ror.org/01h0zpd94"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":30,"referenced_works":["https://openalex.org/W1999970121","https://openalex.org/W2194775991","https://openalex.org/W2765811365","https://openalex.org/W2998218113","https://openalex.org/W3152893301","https://openalex.org/W3172512547","https://openalex.org/W3211787299","https://openalex.org/W4281485151","https://openalex.org/W4281627997","https://openalex.org/W4281661987","https://openalex.org/W4281690218","https://openalex.org/W4308479898","https://openalex.org/W4312933868","https://openalex.org/W4360831816","https://openalex.org/W4362607219","https://openalex.org/W4382239902","https://openalex.org/W4382657733","https://openalex.org/W4386057777","https://openalex.org/W4386065704","https://openalex.org/W4386071957","https://openalex.org/W4386758232","https://openalex.org/W4386845850","https://openalex.org/W4387789586","https://openalex.org/W4390871980","https://openalex.org/W4390872297","https://openalex.org/W4390874074","https://openalex.org/W4390874580","https://openalex.org/W4393578753","https://openalex.org/W6797854001","https://openalex.org/W6838815585"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2111241003","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2096844293","https://openalex.org/W2363944576","https://openalex.org/W2351041855","https://openalex.org/W2570254841"],"abstract_inverted_index":{"The":[0],"diffusion":[1,31,60,78,123,166,201],"model":[2,32,79,89],"has":[3],"achieved":[4],"remarkable":[5],"success":[6],"in":[7,63,222],"the":[8,30,38,42,59,77,81,102,111,117,138,152,156,162,178,185,194,219,254],"era":[9],"of":[10,41,165],"Artificial":[11],"Intelligence":[12],"Generated":[13],"Content":[14],"(AIGC)":[15],"across":[16],"various":[17],"tasks,":[18],"such":[19],"as":[20],"image,":[21],"video,":[22],"text,":[23],"material":[24],"modeling,":[25],"and":[26,168,188,204,211,239,242,247,274,280],"molecular":[27],"design.":[28],"However,":[29],"is":[33,53],"computational":[34,153,191],"intensive":[35],"due":[36],"to":[37,57,75,109,136,150,176,272],"long":[39,179],"iteration":[40,180],"reverse":[43],"denoising":[44],"process,":[45],"which":[46],"hinders":[47],"its":[48],"further":[49],"advancement.":[50],"Therefore,":[51],"there":[52],"an":[54,260],"urgent":[55],"need":[56],"accelerate":[58,76],"model,":[61],"especially":[62],"edge":[64],"scenarios":[65],"that":[66,131,214],"require":[67],"real-time":[68],"computation.":[69],"While":[70],"researchers":[71],"have":[72,100],"made":[73],"efforts":[74],"at":[80,193],"algorithm":[82],"level":[83],"using":[84],"either":[85],"efficient":[86],"sampling":[87],"or":[88],"quantization,":[90],"they":[91,99],"still":[92],"suffer":[93],"from":[94],"accuracy":[95],"degradation.":[96],"More":[97],"importantly,":[98],"overlooked":[101],"hardware-level":[103],"acceleration":[104],"challenge.":[105,181],"This":[106],"work":[107],"aims":[108],"bridge":[110],"gap":[112],"by":[113],"introducing":[114],"Diff-Acc":[115,251,269],",":[116],"first":[118],"FPGA":[119,258],"accelerator":[120],"for":[121],"unconditional":[122,200],"models":[124,167,202],"with":[125,173,206,234],"a":[126,170],"novel":[127,171],"step-wise":[128],"quantization":[129,142,216],"method":[130,217],"requires":[132],"minimal":[133],"calibration":[134],"data":[135,186],"achieve":[137],"state-of-the-art":[139],"(SOTA)":[140],"PTQ":[141,231],"accuracy.":[143],"Additionally,":[144],"we":[145,159,183],"adopt":[146,189],"several":[147],"hardware-oriented":[148],"optimizations":[149],"reduce":[151],"overhead.":[154],"At":[155],"architecture":[157,172],"level,":[158],"fully":[160],"analyze":[161],"computation":[163],"flow":[164],"propose":[169],"group-wise":[174],"parallelism":[175],"tackle":[177],"Besides,":[182],"decouple":[184],"dependencies":[187],"proper":[190],"transformations":[192],"micro-architecture":[195],"level.":[196],"Experiments":[197],"on":[198,253],"two":[199,207],"(DDIM":[203],"DDPM)":[205],"image":[208,223],"datasets":[209],"(CIFAR-10":[210],"ImageNet)":[212],"demonstrate":[213],"our":[215],"achieves":[218,270],"substantial":[220],"improvements":[221,277],"quality":[224],"(FID:":[225],"6.67,":[226],"sFID:":[227],"11.24)":[228],"under":[229],"8-bit":[230],"quantization.":[232],"Compared":[233],"both":[235],"server-based":[236],"(Tesla":[237],"V100":[238],"Intel":[240],"Xeon)":[241],"edge-based":[243,267],"(Raspberry":[244],"Pi":[245],"4":[246],"Jetson":[248],"Nano)":[249],"platforms,":[250,268],"implemented":[252],"Zynq":[255],"UltraScale+":[256],"XCZU9EG":[257],"demonstrates":[259],"up-to":[261],"12.5\u00d7":[262],"energy":[263],"efficiency.":[264],"Particularly":[265],"versus":[266],"up":[271],"10.26\u00d7":[273],"1.97\u00d7":[275],"performance":[276],"over":[278],"CPU":[279],"GPU,":[281],"respectively.":[282]},"counts_by_year":[{"year":2026,"cited_by_count":2}],"updated_date":"2026-05-21T06:26:12.895304","created_date":"2025-10-10T00:00:00"}