{"id":"https://openalex.org/W4413308156","doi":"https://doi.org/10.1145/3762190","title":"eMamba: Efficient Acceleration Framework for Mamba Models in Edge Computing","display_name":"eMamba: Efficient Acceleration Framework for Mamba Models in Edge Computing","publication_year":2025,"publication_date":"2025-08-19","ids":{"openalex":"https://openalex.org/W4413308156","doi":"https://doi.org/10.1145/3762190"},"language":"en","primary_location":{"id":"doi:10.1145/3762190","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3762190","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3762190","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":true,"oa_status":"bronze","oa_url":"https://dl.acm.org/doi/pdf/10.1145/3762190","any_repository_has_fulltext":false},"authorships":[{"author_position":"first","author":{"id":null,"display_name":"Jiyong Kim","orcid":"https://orcid.org/0009-0000-6857-755X"},"institutions":[{"id":"https://openalex.org/I40542001","display_name":"University of Ulsan","ror":"https://ror.org/02c2f8975","country_code":"KR","type":"education","lineage":["https://openalex.org/I40542001"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jiyong Kim","raw_affiliation_strings":["Department of Electrical, Electronic and Computer Engineering, University of Ulsan"],"raw_orcid":"https://orcid.org/0009-0000-6857-755X","affiliations":[{"raw_affiliation_string":"Department of Electrical, Electronic and Computer Engineering, University of Ulsan","institution_ids":["https://openalex.org/I40542001"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039373039","display_name":"J Y Lee","orcid":null},"institutions":[{"id":"https://openalex.org/I40542001","display_name":"University of Ulsan","ror":"https://ror.org/02c2f8975","country_code":"KR","type":"education","lineage":["https://openalex.org/I40542001"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jaeho Lee","raw_affiliation_strings":["Department of Electrical, Electronic and Computer Engineering, University of Ulsan"],"raw_orcid":"https://orcid.org/0009-0009-5434-7791","affiliations":[{"raw_affiliation_string":"Department of Electrical, Electronic and Computer Engineering, University of Ulsan","institution_ids":["https://openalex.org/I40542001"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5109916940","display_name":"Jiun-Shiung Lin","orcid":"https://orcid.org/0009-0000-3618-2385"},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Jiahao Lin","raw_affiliation_strings":["University of Wisconsin-Madison"],"raw_orcid":"https://orcid.org/0009-0000-3618-2385","affiliations":[{"raw_affiliation_string":"University of Wisconsin-Madison","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5006318342","display_name":"Alish Kanani","orcid":"https://orcid.org/0009-0000-8585-9241"},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Alish Kanani","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Wisconsin-Madison"],"raw_orcid":"https://orcid.org/0009-0000-8585-9241","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Wisconsin-Madison","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5008871630","display_name":"Miao Sun","orcid":"https://orcid.org/0000-0002-4537-6998"},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Sun Miao","raw_affiliation_strings":["University of Wisconsin-Madison"],"raw_orcid":"https://orcid.org/0000-0002-4537-6998","affiliations":[{"raw_affiliation_string":"University of Wisconsin-Madison","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5084255924","display_name":"\u00dcmit Y. Ogras","orcid":"https://orcid.org/0000-0002-5045-5535"},"institutions":[{"id":"https://openalex.org/I135310074","display_name":"University of Wisconsin\u2013Madison","ror":"https://ror.org/01y2jtd41","country_code":"US","type":"education","lineage":["https://openalex.org/I135310074"]}],"countries":["US"],"is_corresponding":false,"raw_author_name":"Umit Ogras","raw_affiliation_strings":["Department of Electrical and Computer Engineering, University of Wisconsin-Madison"],"raw_orcid":"https://orcid.org/0000-0002-5045-5535","affiliations":[{"raw_affiliation_string":"Department of Electrical and Computer Engineering, University of Wisconsin-Madison","institution_ids":["https://openalex.org/I135310074"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5100618955","display_name":"Jaehyun Park","orcid":"https://orcid.org/0000-0002-2276-4998"},"institutions":[{"id":"https://openalex.org/I40542001","display_name":"University of Ulsan","ror":"https://ror.org/02c2f8975","country_code":"KR","type":"education","lineage":["https://openalex.org/I40542001"]}],"countries":["KR"],"is_corresponding":false,"raw_author_name":"Jaehyun Park","raw_affiliation_strings":["Department of Electrical, Electronic and Computer Engineering, University of Ulsan"],"raw_orcid":"https://orcid.org/0000-0002-2276-4998","affiliations":[{"raw_affiliation_string":"Department of Electrical, Electronic and Computer Engineering, University of Ulsan","institution_ids":["https://openalex.org/I40542001"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":7,"corresponding_author_ids":[],"corresponding_institution_ids":[],"apc_list":null,"apc_paid":null,"fwci":6.9512,"has_fulltext":true,"cited_by_count":4,"citation_normalized_percentile":{"value":0.96970005,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":"24","issue":"5s","first_page":"1","last_page":"22"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10054","display_name":"Parallel Computing and Optimization Techniques","score":0.9923999905586243,"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":0.9923999905586243,"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/T10444","display_name":"Context-Aware Activity Recognition Systems","score":0.9829000234603882,"subfield":{"id":"https://openalex.org/subfields/1707","display_name":"Computer Vision and Pattern Recognition"},"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":0.9684000015258789,"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/acceleration","display_name":"Acceleration","score":0.7623998522758484},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.525107204914093},{"id":"https://openalex.org/keywords/enhanced-data-rates-for-gsm-evolution","display_name":"Enhanced Data Rates for GSM Evolution","score":0.5171958208084106},{"id":"https://openalex.org/keywords/physics","display_name":"Physics","score":0.27294033765792847},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.10650292038917542},{"id":"https://openalex.org/keywords/classical-mechanics","display_name":"Classical mechanics","score":0.051911234855651855}],"concepts":[{"id":"https://openalex.org/C117896860","wikidata":"https://www.wikidata.org/wiki/Q11376","display_name":"Acceleration","level":2,"score":0.7623998522758484},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.525107204914093},{"id":"https://openalex.org/C162307627","wikidata":"https://www.wikidata.org/wiki/Q204833","display_name":"Enhanced Data Rates for GSM Evolution","level":2,"score":0.5171958208084106},{"id":"https://openalex.org/C121332964","wikidata":"https://www.wikidata.org/wiki/Q413","display_name":"Physics","level":0,"score":0.27294033765792847},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.10650292038917542},{"id":"https://openalex.org/C74650414","wikidata":"https://www.wikidata.org/wiki/Q11397","display_name":"Classical mechanics","level":1,"score":0.051911234855651855}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1145/3762190","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3762190","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3762190","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"}],"best_oa_location":{"id":"doi:10.1145/3762190","is_oa":true,"landing_page_url":"https://doi.org/10.1145/3762190","pdf_url":"https://dl.acm.org/doi/pdf/10.1145/3762190","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"},"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G8604652154","display_name":null,"funder_award_id":"RS-2023-00208046","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/W4413308156.pdf","grobid_xml":"https://content.openalex.org/works/W4413308156.grobid-xml"},"referenced_works_count":23,"referenced_works":["https://openalex.org/W179875071","https://openalex.org/W2442974303","https://openalex.org/W2565125333","https://openalex.org/W2603836393","https://openalex.org/W2891946740","https://openalex.org/W2982479999","https://openalex.org/W3004127905","https://openalex.org/W3135539146","https://openalex.org/W3201387271","https://openalex.org/W3201448882","https://openalex.org/W3211787299","https://openalex.org/W4281562926","https://openalex.org/W4308090791","https://openalex.org/W4318767343","https://openalex.org/W4360831795","https://openalex.org/W4385187240","https://openalex.org/W4385326807","https://openalex.org/W4387494819","https://openalex.org/W4390873361","https://openalex.org/W4398233154","https://openalex.org/W4406266037","https://openalex.org/W6797854001","https://openalex.org/W6968655253"],"related_works":["https://openalex.org/W4391375266","https://openalex.org/W2899084033","https://openalex.org/W2748952813","https://openalex.org/W2390279801","https://openalex.org/W4391913857","https://openalex.org/W2358668433","https://openalex.org/W4396701345","https://openalex.org/W2376932109","https://openalex.org/W2001405890","https://openalex.org/W4396696052"],"abstract_inverted_index":{"State":[0],"Space":[1],"Model":[2],"(SSM)-based":[3],"machine":[4],"learning":[5],"architectures":[6],"have":[7],"recently":[8],"gained":[9],"significant":[10],"attention":[11],"for":[12,40,51,69],"processing":[13],"sequential":[14],"data.":[15],"Mamba,":[16],"a":[17,61],"recent":[18],"sequence-to-sequence":[19],"SSM,":[20],"offers":[21],"competitive":[22,214],"accuracy":[23,136],"with":[24,85,121,197],"superior":[25],"computational":[26,78],"efficiency":[27,79],"compared":[28],"to":[29,112,137,149],"state-of-the-art":[30,138],"transformer":[31],"models.":[32],"While":[33],"this":[34],"advantage":[35],"makes":[36],"Mamba":[37,71],"particularly":[38],"promising":[39],"resource-constrained":[41],"edge":[42,74],"devices,":[43],"no":[44],"hardware":[45,64],"acceleration":[46,65],"frameworks":[47],"are":[48],"currently":[49],"optimized":[50],"deploying":[52,70],"it":[53,104,146],"in":[54],"such":[55,93],"environments.":[56],"This":[57],"article":[58],"presents":[59],"eMamba,":[60],"comprehensive":[62],"end-to-end":[63],"framework":[66],"explicitly":[67],"designed":[68],"models":[72],"on":[73,161,174],"platforms.":[75],"eMamba":[76,133,172],"maximizes":[77],"by":[80],"replacing":[81],"complex":[82],"normalization":[83],"layers":[84],"lightweight":[86],"hardware-aware":[87],"alternatives":[88],"and":[89,97,124,168,179,193,204],"approximating":[90],"expensive":[91],"operations,":[92],"as":[94],"SiLU":[95],"activation":[96],"exponentiation,":[98],"considering":[99],"the":[100,114,162,170],"target":[101],"applications.":[102],"Then,":[103],"performs":[105],"an":[106,126,175],"approximation-aware":[107],"neural":[108],"architecture":[109],"search":[110],"(NAS)":[111],"tune":[113],"learnable":[115],"parameters":[116],"used":[117],"during":[118],"approximation.":[119],"Evaluations":[120],"Fashion-MNIST,":[122],"CIFAR-10,":[123],"MARS,":[125],"open-source":[127],"human":[128],"pose":[129],"estimation":[130],"dataset,":[131],"show":[132,189],"achieves":[134],"comparable":[135],"techniques":[139],"using":[140,181],"1.63\u201319.9\u00d7":[141],"fewer":[142],"parameters.":[143],"In":[144],"addition,":[145],"generalizes":[147],"well":[148],"large-scale":[150],"natural":[151],"language":[152],"tasks,":[153],"demonstrating":[154],"stable":[155],"perplexity":[156],"across":[157],"varying":[158],"sequence":[159],"lengths":[160],"WikiText2":[163],"dataset.":[164],"We":[165],"also":[166],"quantize":[167],"implement":[169],"entire":[171],"pipeline":[173],"AMD":[176],"ZCU102":[177],"FPGA":[178],"ASIC":[180],"GlobalFoundries":[182],"(GF)":[183],"22":[184],"nm":[185],"technology.":[186],"Experimental":[187],"results":[188],"4.95\u20135.62\u00d7":[190],"lower":[191,202,206],"latency":[192],"2.22\u20139.95\u00d7":[194],"higher":[195],"throughput,":[196],"4.77\u00d7":[198],"smaller":[199],"area,":[200],"9.84\u00d7":[201],"power,":[203],"48.6\u00d7":[205],"energy":[207],"consumption":[208],"than":[209],"baseline":[210],"solutions":[211],"while":[212],"maintaining":[213],"accuracy.":[215]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":1}],"updated_date":"2026-06-11T09:08:48.828518","created_date":"2025-10-10T00:00:00"}
