{"id":"https://openalex.org/W4413947071","doi":"https://doi.org/10.1109/tc.2025.3604473","title":"CoFormer: Collaborating With Heterogeneous Edge Devices for Scalable Transformer Inference","display_name":"CoFormer: Collaborating With Heterogeneous Edge Devices for Scalable Transformer Inference","publication_year":2025,"publication_date":"2025-09-02","ids":{"openalex":"https://openalex.org/W4413947071","doi":"https://doi.org/10.1109/tc.2025.3604473"},"language":"en","primary_location":{"id":"doi:10.1109/tc.2025.3604473","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tc.2025.3604473","pdf_url":null,"source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"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 Computers","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/A5047347817","display_name":"Guanyu Xu","orcid":"https://orcid.org/0000-0002-2771-9272"},"institutions":[{"id":"https://openalex.org/I125839683","display_name":"Beijing Institute of Technology","ror":"https://ror.org/01skt4w74","country_code":"CN","type":"education","lineage":["https://openalex.org/I125839683","https://openalex.org/I890469752"]}],"countries":["CN"],"is_corresponding":true,"raw_author_name":"Guanyu Xu","raw_affiliation_strings":["School of Information and Electronics, Beijing Institute of Technology, Beijing, China","School of Information and Electrionics, Beijing Institute of Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-2771-9272","affiliations":[{"raw_affiliation_string":"School of Information and Electronics, Beijing Institute of Technology, Beijing, China","institution_ids":["https://openalex.org/I125839683"]},{"raw_affiliation_string":"School of Information and Electrionics, Beijing Institute of Technology, Beijing, China","institution_ids":["https://openalex.org/I125839683"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5057198274","display_name":"Zhiwei Hao","orcid":"https://orcid.org/0000-0002-6237-7028"},"institutions":[{"id":"https://openalex.org/I125839683","display_name":"Beijing Institute of Technology","ror":"https://ror.org/01skt4w74","country_code":"CN","type":"education","lineage":["https://openalex.org/I125839683","https://openalex.org/I890469752"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Zhiwei Hao","raw_affiliation_strings":["School of Information and Electronics, Beijing Institute of Technology, Beijing, China","School of Information and Electrionics, Beijing Institute of Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0002-6237-7028","affiliations":[{"raw_affiliation_string":"School of Information and Electronics, Beijing Institute of Technology, Beijing, China","institution_ids":["https://openalex.org/I125839683"]},{"raw_affiliation_string":"School of Information and Electrionics, Beijing Institute of Technology, Beijing, China","institution_ids":["https://openalex.org/I125839683"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5042802004","display_name":"Li Shen","orcid":"https://orcid.org/0000-0001-5659-3464"},"institutions":[{"id":"https://openalex.org/I157773358","display_name":"Sun Yat-sen University","ror":"https://ror.org/0064kty71","country_code":"CN","type":"education","lineage":["https://openalex.org/I157773358"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Li Shen","raw_affiliation_strings":["School of Cyber Science and Technology, Shenzhen Campus of Sun Yat-sen University, Shenzhen, China"],"raw_orcid":"https://orcid.org/0000-0001-5659-3464","affiliations":[{"raw_affiliation_string":"School of Cyber Science and Technology, Shenzhen Campus of Sun Yat-sen University, Shenzhen, China","institution_ids":["https://openalex.org/I157773358"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5039158745","display_name":"Yong Luo","orcid":"https://orcid.org/0000-0002-2296-6370"},"institutions":[{"id":"https://openalex.org/I37461747","display_name":"Wuhan University","ror":"https://ror.org/033vjfk17","country_code":"CN","type":"education","lineage":["https://openalex.org/I37461747"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Yong Luo","raw_affiliation_strings":["School of Computer Science, National Engineering Research Center for Multimedia Software, Wuhan University, Wuhan, China"],"raw_orcid":"https://orcid.org/0000-0002-2296-6370","affiliations":[{"raw_affiliation_string":"School of Computer Science, National Engineering Research Center for Multimedia Software, Wuhan University, Wuhan, China","institution_ids":["https://openalex.org/I37461747"]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5041332590","display_name":"Fuhui Sun","orcid":"https://orcid.org/0009-0004-5341-0158"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Fuhui Sun","raw_affiliation_strings":["Information Technology Service Center of People’s Court, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0004-5341-0158","affiliations":[{"raw_affiliation_string":"Information Technology Service Center of People’s Court, Beijing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":null,"display_name":"Xiaoyan Wang","orcid":"https://orcid.org/0009-0000-3972-1575"},"institutions":[],"countries":[],"is_corresponding":false,"raw_author_name":"Xiaoyan Wang","raw_affiliation_strings":["Information Technology Service Center of People’s Court, Beijing, China"],"raw_orcid":"https://orcid.org/0009-0000-3972-1575","affiliations":[{"raw_affiliation_string":"Information Technology Service Center of People’s Court, Beijing, China","institution_ids":[]}]},{"author_position":"middle","author":{"id":"https://openalex.org/A5020194895","display_name":"Han Hu","orcid":"https://orcid.org/0000-0001-7532-0496"},"institutions":[{"id":"https://openalex.org/I125839683","display_name":"Beijing Institute of Technology","ror":"https://ror.org/01skt4w74","country_code":"CN","type":"education","lineage":["https://openalex.org/I125839683","https://openalex.org/I890469752"]}],"countries":["CN"],"is_corresponding":false,"raw_author_name":"Han Hu","raw_affiliation_strings":["School of Information and Electronics, Beijing Institute of Technology, Beijing, China","School of Information and Electrionics, Beijing Institute of Technology, Beijing, China"],"raw_orcid":"https://orcid.org/0000-0001-7532-0496","affiliations":[{"raw_affiliation_string":"School of Information and Electronics, Beijing Institute of Technology, Beijing, China","institution_ids":["https://openalex.org/I125839683"]},{"raw_affiliation_string":"School of Information and Electrionics, Beijing Institute of Technology, Beijing, China","institution_ids":["https://openalex.org/I125839683"]}]},{"author_position":"last","author":{"id":"https://openalex.org/A5041572550","display_name":"Yonggang Wen","orcid":"https://orcid.org/0000-0002-2751-5114"},"institutions":[{"id":"https://openalex.org/I172675005","display_name":"Nanyang Technological University","ror":"https://ror.org/02e7b5302","country_code":"SG","type":"education","lineage":["https://openalex.org/I172675005"]}],"countries":["SG"],"is_corresponding":false,"raw_author_name":"Yonggang Wen","raw_affiliation_strings":["College of Computing and Data Science, Nanyang Technological University, Singapore"],"raw_orcid":"https://orcid.org/0000-0002-2751-5114","affiliations":[{"raw_affiliation_string":"College of Computing and Data Science, Nanyang Technological University, Singapore","institution_ids":["https://openalex.org/I172675005"]}]}],"institutions":[],"countries_distinct_count":2,"institutions_distinct_count":8,"corresponding_author_ids":["https://openalex.org/A5047347817"],"corresponding_institution_ids":["https://openalex.org/I125839683"],"apc_list":null,"apc_paid":null,"fwci":0.0,"has_fulltext":false,"cited_by_count":0,"citation_normalized_percentile":{"value":0.20965028,"is_in_top_1_percent":false,"is_in_top_10_percent":false},"cited_by_percentile_year":null,"biblio":{"volume":"74","issue":"12","first_page":"4010","last_page":"4024"},"is_retracted":false,"is_paratext":false,"is_xpac":false,"primary_topic":{"id":"https://openalex.org/T10369","display_name":"Advanced MEMS and NEMS Technologies","score":0.9398999810218811,"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"}},"topics":[{"id":"https://openalex.org/T10369","display_name":"Advanced MEMS and NEMS Technologies","score":0.9398999810218811,"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/T12611","display_name":"Neural Networks and Reservoir Computing","score":0.9362000226974487,"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/T12452","display_name":"Electrowetting and Microfluidic Technologies","score":0.9180999994277954,"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/computer-science","display_name":"Computer science","score":0.69724041223526},{"id":"https://openalex.org/keywords/inference","display_name":"Inference","score":0.5988163352012634},{"id":"https://openalex.org/keywords/scalability","display_name":"Scalability","score":0.5944080352783203},{"id":"https://openalex.org/keywords/transformer","display_name":"Transformer","score":0.42411500215530396},{"id":"https://openalex.org/keywords/computer-architecture","display_name":"Computer architecture","score":0.3965147137641907},{"id":"https://openalex.org/keywords/parallel-computing","display_name":"Parallel computing","score":0.348410427570343},{"id":"https://openalex.org/keywords/artificial-intelligence","display_name":"Artificial intelligence","score":0.18430495262145996},{"id":"https://openalex.org/keywords/electrical-engineering","display_name":"Electrical engineering","score":0.1565842628479004},{"id":"https://openalex.org/keywords/engineering","display_name":"Engineering","score":0.12750890851020813},{"id":"https://openalex.org/keywords/operating-system","display_name":"Operating system","score":0.11718478798866272},{"id":"https://openalex.org/keywords/voltage","display_name":"Voltage","score":0.11356258392333984}],"concepts":[{"id":"https://openalex.org/C41008148","wikidata":"https://www.wikidata.org/wiki/Q21198","display_name":"Computer science","level":0,"score":0.69724041223526},{"id":"https://openalex.org/C2776214188","wikidata":"https://www.wikidata.org/wiki/Q408386","display_name":"Inference","level":2,"score":0.5988163352012634},{"id":"https://openalex.org/C48044578","wikidata":"https://www.wikidata.org/wiki/Q727490","display_name":"Scalability","level":2,"score":0.5944080352783203},{"id":"https://openalex.org/C66322947","wikidata":"https://www.wikidata.org/wiki/Q11658","display_name":"Transformer","level":3,"score":0.42411500215530396},{"id":"https://openalex.org/C118524514","wikidata":"https://www.wikidata.org/wiki/Q173212","display_name":"Computer architecture","level":1,"score":0.3965147137641907},{"id":"https://openalex.org/C173608175","wikidata":"https://www.wikidata.org/wiki/Q232661","display_name":"Parallel computing","level":1,"score":0.348410427570343},{"id":"https://openalex.org/C154945302","wikidata":"https://www.wikidata.org/wiki/Q11660","display_name":"Artificial intelligence","level":1,"score":0.18430495262145996},{"id":"https://openalex.org/C119599485","wikidata":"https://www.wikidata.org/wiki/Q43035","display_name":"Electrical engineering","level":1,"score":0.1565842628479004},{"id":"https://openalex.org/C127413603","wikidata":"https://www.wikidata.org/wiki/Q11023","display_name":"Engineering","level":0,"score":0.12750890851020813},{"id":"https://openalex.org/C111919701","wikidata":"https://www.wikidata.org/wiki/Q9135","display_name":"Operating system","level":1,"score":0.11718478798866272},{"id":"https://openalex.org/C165801399","wikidata":"https://www.wikidata.org/wiki/Q25428","display_name":"Voltage","level":2,"score":0.11356258392333984}],"mesh":[],"locations_count":1,"locations":[{"id":"doi:10.1109/tc.2025.3604473","is_oa":false,"landing_page_url":"https://doi.org/10.1109/tc.2025.3604473","pdf_url":null,"source":{"id":"https://openalex.org/S157670870","display_name":"IEEE Transactions on Computers","issn_l":"0018-9340","issn":["0018-9340","1557-9956","2326-3814"],"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 Computers","raw_type":"journal-article"}],"best_oa_location":null,"sustainable_development_goals":[],"awards":[{"id":"https://openalex.org/G5019704030","display_name":null,"funder_award_id":"U2336211","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G5548306916","display_name":null,"funder_award_id":"92467206","funder_id":"https://openalex.org/F4320321001","funder_display_name":"National Natural Science Foundation of China"},{"id":"https://openalex.org/G8276535520","display_name":null,"funder_award_id":"U23A20318","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"}],"has_content":{"pdf":false,"grobid_xml":false},"content_urls":null,"referenced_works_count":33,"referenced_works":["https://openalex.org/W2896180420","https://openalex.org/W2896457183","https://openalex.org/W2946794439","https://openalex.org/W2963150697","https://openalex.org/W2963420686","https://openalex.org/W2972324944","https://openalex.org/W2983488099","https://openalex.org/W3094502228","https://openalex.org/W3110777925","https://openalex.org/W3121523901","https://openalex.org/W3138516171","https://openalex.org/W3173381523","https://openalex.org/W3175515048","https://openalex.org/W3210764291","https://openalex.org/W4281554877","https://openalex.org/W4285217780","https://openalex.org/W4285261368","https://openalex.org/W4292347911","https://openalex.org/W4293195477","https://openalex.org/W4306178486","https://openalex.org/W4308112567","https://openalex.org/W4312191382","https://openalex.org/W4312446817","https://openalex.org/W4312516520","https://openalex.org/W4312820606","https://openalex.org/W4367670059","https://openalex.org/W4386071870","https://openalex.org/W4386072014","https://openalex.org/W4386699354","https://openalex.org/W4401508667","https://openalex.org/W4401568153","https://openalex.org/W4405778753","https://openalex.org/W4405934565"],"related_works":["https://openalex.org/W2389214306","https://openalex.org/W2965083567","https://openalex.org/W4235240664","https://openalex.org/W1838576100","https://openalex.org/W2095886385","https://openalex.org/W4401278057","https://openalex.org/W4206178588","https://openalex.org/W3094491777","https://openalex.org/W3214715529","https://openalex.org/W4287635093"],"abstract_inverted_index":{"The":[0,92],"impressive":[1],"performance":[2],"of":[3,10,38,104,181,204],"transformer":[4,46,88,109,182,196],"models":[5,55,116,168,183],"has":[6],"sparked":[7],"the":[8,31,100,128,155,160,174,201],"deployment":[9],"intelligent":[11],"applications":[12],"on":[13,56,184,210],"resource-constrained":[14],"edge":[15,23,58,186,211],"devices.":[16,59],"However,":[17],"ensuring":[18],"high-quality":[19],"service":[20],"for":[21,86,117],"real-time":[22],"systems":[24],"is":[25,97,150],"a":[26,82,178],"significant":[27],"challenge":[28],"due":[29],"to":[30,48,98,126,136,152,158,169,176,190],"considerable":[32,66],"computational":[33],"demands":[34],"and":[35,74,102,120,141,163],"resource":[36],"requirements":[37,215],"these":[39,78],"models.":[40,197],"Existing":[41],"strategies":[42],"typically":[43],"either":[44,65],"offload":[45],"computations":[47],"other":[49],"devices":[50],"or":[51,69],"directly":[52],"deploy":[53],"compressed":[54],"individual":[57],"These":[60],"strategies,":[61],"however,":[62],"result":[63],"in":[64],"communication":[67],"overhead":[68],"suboptimal":[70],"trade-offs":[71],"between":[72],"accuracy":[73,142],"efficiency.":[75],"To":[76],"tackle":[77],"challenges,":[79],"we":[80],"propose":[81],"collaborative":[83],"inference":[84,139,192,203,230],"system":[85],"general":[87],"models,":[89],"termed":[90],"CoFormer.":[91],"central":[93],"idea":[94],"behind":[95],"CoFormer":[96,199,218],"exploit":[99],"divisibility":[101],"integrability":[103],"transformer.":[105],"An":[106],"off-the-shelf":[107],"large":[108,195],"can":[110,219],"be":[111],"decomposed":[112,167],"into":[113],"multiple":[114],"smaller":[115],"distributed":[118],"inference,":[119],"their":[121],"intermediate":[122],"results":[123],"are":[124],"aggregated":[125],"generate":[127],"final":[129],"output.":[130],"We":[131,172],"formulate":[132],"an":[133],"optimization":[134,156],"problem":[135,157],"minimize":[137],"both":[138],"latency":[140],"degradation":[143],"under":[144],"heterogeneous":[145,185],"hardware":[146],"constraints.":[147],"DeBo":[148],"algorithm":[149],"proposed":[151],"first":[153],"solve":[154],"derive":[159],"decomposition":[161],"policy,":[162],"then":[164],"progressively":[165],"calibrate":[166],"restore":[170],"performance.":[171,231],"demonstrate":[173],"capability":[175],"support":[177],"wide":[179],"range":[180],"devices,":[187,212],"achieving":[188],"up":[189],"3.1\u00d7":[191],"speedup":[193],"with":[194,206],"Notably,":[198],"enables":[200],"efficient":[202],"GPT2-XL":[205],"1.6":[207],"billion":[208],"parameters":[209],"reducing":[213],"memory":[214],"by":[216,224],"76.3%.":[217],"also":[220],"reduce":[221],"energy":[222],"consumption":[223],"approximately":[225],"40%":[226],"while":[227],"maintaining":[228],"satisfactory":[229]},"counts_by_year":[],"updated_date":"2026-04-29T09:16:38.111599","created_date":"2025-10-10T00:00:00"}