{"id":"https://openalex.org/W4409723502","doi":"https://doi.org/10.1109/tai.2025.3563144","title":"Efficient Transformer Inference Through Hybrid Dynamic Pruning","display_name":"Efficient Transformer Inference Through Hybrid Dynamic Pruning","publication_year":2025,"publication_date":"2025-04-23","ids":{"openalex":"https://openalex.org/W4409723502","doi":"https://doi.org/10.1109/tai.2025.3563144"},"language":"en","primary_location":{"id":"doi:10.1109/tai.2025.3563144","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tai.2025.3563144","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","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/A5054277953","display_name":"Ghadeer A. Jaradat","orcid":null},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Ghadeer A. Jaradat","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","SoC Center, Khalifa University, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0009-0003-7543-6781","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5091151249","display_name":"Mohammed F. Tolba","orcid":"https://orcid.org/0000-0002-6412-290X"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Mohammed F. Tolba","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","SoC Center, Khalifa University, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-6412-290X","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039488421","display_name":"Ghada Alsuhli","orcid":"https://orcid.org/0000-0001-7251-747X"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Ghada Alsuhli","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","SoC Center, Khalifa University, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0001-7251-747X","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5078108496","display_name":"Hani Saleh","orcid":"https://orcid.org/0000-0002-7185-0278"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Hani Saleh","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","SoC Center, Khalifa University, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-7185-0278","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020682177","display_name":"Mahmoud Al\u2010Qutayri","orcid":"https://orcid.org/0000-0002-9600-8036"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Mahmoud Al-Qutayri","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","SoC Center, Khalifa University, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-9600-8036","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5045307151","display_name":"T. Stouraitis","orcid":"https://orcid.org/0000-0002-3696-4958"},"institutions":[{"id":"https://openalex.org/I176601375","display_name":"Khalifa University of Science and Technology","ror":"https://ror.org/05hffr360","country_code":"AE","type":"education","lineage":["https://openalex.org/I176601375"]}],"countries":["AE"],"is_corresponding":false,"raw_author_name":"Thanos Stouraitis","raw_affiliation_strings":["SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","SoC Center, Khalifa University, Abu Dhabi, UAE"],"raw_orcid":"https://orcid.org/0000-0002-3696-4958","affiliations":[{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, United Arab Emirates","institution_ids":["https://openalex.org/I176601375"]},{"raw_affiliation_string":"SoC Center, Khalifa University, Abu Dhabi, UAE","institution_ids":["https://openalex.org/I176601375"]}]}],"institutions":[],"countries_distinct_count":1,"institutions_distinct_count":1,"corresponding_author_ids":[],"corresponding_institution_ids":["https://openalex.org/I176601375"],"apc_list":null,"apc_paid":null,"fwci":5.584,"has_fulltext":false,"cited_by_count":4,"citation_normalized_percentile":{"value":0.95328705,"is_in_top_1_percent":false,"is_in_top_10_percent":true},"cited_by_percentile_year":{"min":91,"max":99},"biblio":{"volume":"6","issue":"12","first_page":"3273","last_page":"3286"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.7440000176429749,"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"}},"topics":[{"id":"https://openalex.org/T10320","display_name":"Neural Networks and Applications","score":0.7440000176429749,"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/T11343","display_name":"Power Transformer Diagnostics and Insulation","score":0.6740999817848206,"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/T12169","display_name":"Non-Destructive Testing Techniques","score":0.6453999876976013,"subfield":{"id":"https://openalex.org/subfields/2210","display_name":"Mechanical 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/inference","display_name":"Inference","score":0.7207031846046448},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.5805355310440063},{"id":"https://openalex.org/keywords/computer-science","display_name":"Computer science","score":0.4620904326438904},{"id":"https://openalex.org/keywords/pruning","display_name":"Pruning","score":0.4260774254798889},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.4072189927101135},{"id":"https://openalex.org/keywords/machine-learning","display_name":"Machine learning","score":0.3241952955722809},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.11439689993858337},{"id":"https://openalex.org/keywords/biology","display_name":"Biology","score":0.10231584310531616},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.0738484263420105},{"id":"https://openalex.org/keywords/botany","display_name":"Botany","score":0.06205272674560547}],"concepts":[{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.7207031846046448},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.5805355310440063},{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.4620904326438904},{"id":"https://openalex.org/C108010975","wikidata":"https://www.wikidata.org/wiki/Q500094","display_name":"Pruning","level":2,"score":0.4260774254798889},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.4072189927101135},{"id":"https://openalex.org/C119857082","wikidata":"https://www.wikidata.org/wiki/Q2539","display_name":"Machine learning","level":1,"score":0.3241952955722809},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.11439689993858337},{"id":"https://openalex.org/C86803240","wikidata":"https://www.wikidata.org/wiki/Q420","display_name":"Biology","level":0,"score":0.10231584310531616},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.0738484263420105},{"id":"https://openalex.org/C59822182","wikidata":"https://www.wikidata.org/wiki/Q441","display_name":"Botany","level":1,"score":0.06205272674560547},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.0}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tai.2025.3563144","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tai.2025.3563144","pdf_url":null,"source":{"id":"https://openalex.org/S4210169448","display_name":"IEEE Transactions on Artificial Intelligence","issn_l":"2691-4581","issn":["2691-4581"],"is_oa":false,"is_in_doaj":false,"is_core":true,"host_organization":"https://openalex.org/P4310319808","host_organization_name":"Institute of Electrical and Electronics Engineers","host_organization_lineage":["https://openalex.org/P4310319808"],"host_organization_lineage_names":["Institute of Electrical and Electronics Engineers"],"type":"journal"},"license":null,"license_id":null,"version":"publishedVersion","is_accepted":true,"is_published":true,"raw_source_name":"IEEE Transactions on Artificial Intelligence","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[{"id":"https://metadata.un.org/sdg/7","display_name":"Affordable and clean energy","score":0.7799999713897705}],"awards":[],"funders":[],"has_content":{"grobid_xml":false,"pdf":false},"content_urls":null,"referenced_works_count":25,"referenced_works":["https://openalex.org/W2923014074","https://openalex.org/W2946794439","https://openalex.org/W2971033911","https://openalex.org/W2997929983","https://openalex.org/W3016542674","https://openalex.org/W3017024317","https://openalex.org/W3034999214","https://openalex.org/W3085139254","https://openalex.org/W3096609285","https://openalex.org/W3106298483","https://openalex.org/W3159727696","https://openalex.org/W3162542754","https://openalex.org/W3166574921","https://openalex.org/W3173222967","https://openalex.org/W3175875420","https://openalex.org/W3190761184","https://openalex.org/W3205192296","https://openalex.org/W3206453033","https://openalex.org/W4214686755","https://openalex.org/W4291653336","https://openalex.org/W4308083526","https://openalex.org/W4312950730","https://openalex.org/W4375928954","https://openalex.org/W4385245566","https://openalex.org/W4386857706"],"related_works":["https://openalex.org/W2961085424","https://openalex.org/W4306674287","https://openalex.org/W4387369504","https://openalex.org/W4394896187","https://openalex.org/W3170094116","https://openalex.org/W4386462264","https://openalex.org/W3107602296","https://openalex.org/W4364306694","https://openalex.org/W4312192474","https://openalex.org/W4283697347"],"abstract_inverted_index":{"In":[0],"the":[1,33,94,97,118,163,178,242],"world":[2],"of":[3,28,35,96,180],"deep":[4],"learning,":[5],"transformer":[6,189],"models":[7,37,190],"have":[8],"become":[9],"very":[10],"significant,":[11],"leading":[12],"to":[13,21,49,84,113,130,215,222,227,234,241],"improvements":[14],"in":[15,38,87,100,117,172,201],"many":[16],"areas":[17],"from":[18],"understanding":[19],"language":[20],"recognizing":[22],"images,":[23],"covering":[24],"a":[25,62,108,168],"wide":[26],"range":[27],"applications.":[29],"Despite":[30],"their":[31,50],"success,":[32],"deployment":[34],"these":[36,58,156],"real-time":[39],"applications,":[40],"particularly":[41],"on":[42],"edge":[43],"devices,":[44],"poses":[45],"significant":[46],"challenges":[47],"due":[48],"computational":[51,244],"intensity":[52],"and":[53,82,89,103,132,175,182,191,198,209,232],"memory":[54,91],"demands.":[55],"To":[56,153],"overcome":[57],"challenges,":[59],"we":[60,106,144,161],"introduce":[61],"novel":[63,109],"Hybrid":[64],"Dynamic":[65],"Pruning":[66],"(HDP)":[67],"technique,":[68],"an":[69,137,146],"efficient":[70],"algorithm-architecture":[71],"co-design":[72],"approach":[73],"that":[74,149,194],"accelerates":[75],"transformers":[76],"using":[77],"head":[78,128],"sparsity,":[79],"block":[80,111],"sparsity":[81],"approximation":[83,147],"reduce":[85,90],"computations":[86],"attention":[88,101,104,119,151,202],"access.":[92],"With":[93],"observation":[95],"huge":[98],"redundancy":[99],"scores":[102],"heads,":[105],"propose":[107,126,145,162],"integer-based":[110],"pruning":[112,129],"prune":[114,133],"unimportant":[115,134],"blocks":[116],"matrix":[120],"at":[121,136,140],"run":[122,141],"time.":[123,142],"We":[124],"also":[125],"integerbased":[127],"detect":[131],"heads":[135],"early":[138],"stage":[139],"Also,":[143],"method":[148],"reduces":[150],"computations.":[152],"efficiently":[154],"support":[155],"methods":[157],"with":[158,187],"lower":[159],"latency,":[160],"HDP":[164],"Accelerator":[165],"(HDPA)":[166],"as":[167],"co-processor":[169],"architecture,":[170],"synthesized":[171],"two":[173],"configurations\u2014HDPA-edge":[174],"HDPA-server\u2014to":[176],"meet":[177],"needs":[179],"mobile":[181],"server":[183],"platforms.":[184],"Extensive":[185],"experiments":[186],"different":[188],"benchmarks":[192],"demonstrate":[193],"HDPA-server":[195],"achieves":[196,220],"481\u00d7":[197],"381\u00d7":[199],"speedup":[200],"layer":[203],"computation":[204],"over":[205],"Intel":[206],"i7-1185G7":[207],"CPU":[208],"NVIDIA":[210],"T4":[211],"GPU,":[212],"respectively.":[213],"Compared":[214],"other":[216],"state-of-the-art":[217],"accelerators,":[218],"HDPA":[219],"1.26\u00d7":[221],"2.08\u00d7":[223],"higher":[224],"throughput,":[225],"1.3\u00d7":[226],"18\u00d7":[228],"greater":[229],"MAC":[230],"efficiency,":[231,238],"1.1\u00d7":[233],"5.1\u00d7":[235],"improved":[236],"energy":[237],"when":[239],"normalized":[240],"same":[243],"load.":[245]},"counts_by_year":[{"year":2026,"cited_by_count":3},{"year":2025,"cited_by_count":1}],"updated_date":"2026-03-27T05:58:40.876381","created_date":"2025-10-10T00:00:00"}
