{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:19:08Z","timestamp":1785543548787,"version":"3.56.0"},"reference-count":75,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"12","license":[{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key R&D Program of China","award":["2021ZD0110400"],"award-info":[{"award-number":["2021ZD0110400"]}]},{"name":"National Key R&D Program of China","award":["2021YFB2900103"],"award-info":[{"award-number":["2021YFB2900103"]}]},{"name":"Innovation Program for Quantum Science and Technology","award":["2021ZD0302900"],"award-info":[{"award-number":["2021ZD0302900"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["623B2093"],"award-info":[{"award-number":["623B2093"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62132018"],"award-info":[{"award-number":["62132018"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62231015"],"award-info":[{"award-number":["62231015"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62441228"],"award-info":[{"award-number":["62441228"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Pioneer and Leading Goose R&D Program of Zhejiang","award":["2023C01029"],"award-info":[{"award-number":["2023C01029"]}]},{"name":"Pioneer and Leading Goose R&D Program of Zhejiang","award":["2023C01143"],"award-info":[{"award-number":["2023C01143"]}]},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["WK2150110024"],"award-info":[{"award-number":["WK2150110024"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Science and Technology Tackling Program of Anhui Province","award":["202423k09020016"],"award-info":[{"award-number":["202423k09020016"]}]},{"name":"RGC Theme-based Research Scheme","award":["2400070"],"award-info":[{"award-number":["2400070"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. on Mobile Comput."],"published-print":{"date-parts":[[2025,12]]},"DOI":"10.1109\/tmc.2025.3588430","type":"journal-article","created":{"date-parts":[[2025,7,11]],"date-time":"2025-07-11T17:46:08Z","timestamp":1752255968000},"page":"13090-13105","source":"Crossref","is-referenced-by-count":2,"title":["Mitigating Tail Latency for On-Device Inference With Load-Balanced Heterogeneous Models"],"prefix":"10.1109","volume":"24","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2624-8755","authenticated-orcid":false,"given":"Mu","family":"Yuan","sequence":"first","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong (CUHK), Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1004-8588","authenticated-orcid":false,"given":"Lan","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, University of Science and Technology of China (USTC), Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4184-0762","authenticated-orcid":false,"given":"Di","family":"Duan","sequence":"additional","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Liekang","family":"Zeng","sequence":"additional","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4280-7406","authenticated-orcid":false,"given":"Miao-Hui","family":"Song","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, University of Science and Technology of China (USTC), Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zichong","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, University of Science and Technology of China (USTC), Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1772-7751","authenticated-orcid":false,"given":"Guoliang","family":"Xing","sequence":"additional","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6070-6625","authenticated-orcid":false,"given":"Xiang-Yang","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, University of Science and Technology of China (USTC), Hefei, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3508396.3512877"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3560905.3568531"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1145\/3495243.3517020"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3643832.3661892"},{"key":"ref5","first-page":"1","article-title":"$\\lbrace${SaFace$\\rbrace$}: Towards scenario-aware face recognition via edge computing system","volume-title":"Proc. 3rd USENIX Workshop Hot Topics Edge Comput.","author":"Zhou"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3636534.3649391"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3447993.3448625"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3300061.3345447"},{"issue":"4","key":"ref9","first-page":"1","article-title":"Mirror: Towards generalizable on-device video virtual try-on for mobile shopping","volume-title":"Proc. ACM Interactive Mobile Wearable Ubiquitous Technol.","volume":"7","author":"Kang","year":"2024"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TCC.2020.3006751"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1145\/3210240.3210313"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.3390\/electronics11060945"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1007\/s10462-020-09816-7"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICDMW.2018.00132"},{"key":"ref15","first-page":"800","article-title":"Tensorflow lite micro: Embedded machine learning for tinyml systems","volume-title":"Proc. Mach. Learn. Syst.","volume":"3","author":"David","year":"2021"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3495243.3517016"},{"key":"ref17","first-page":"359","article-title":"On-camera filtering for resource-efficient real-time video analytics","volume-title":"Proc. Annu. Conf. ACM Special Int. Group Data Commun. Appl. Technol. Architectures, Protoc. Comput. Commun.","author":"Li"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3603269.3604825"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3241539.3241557"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/3230543.3230574"},{"key":"ref21","first-page":"377","article-title":"Live video analytics at scale with approximation and $\\lbrace${Delay-Tolerance$\\rbrace$}","volume-title":"Proc. 14th USENIX Symp. Networked Syst. Des. Implementation","author":"Zhang"},{"key":"ref22","first-page":"1041","article-title":"Cocktail: A multidimensional optimization for model serving in cloud","volume-title":"Proc. 19th USENIX Symp. Networked Syst. Des. Implementation","author":"Gunasekaran"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/3458864.3467884"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3698767"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2018.2816042"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1155\/2018\/6421607"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/TC.1979.1675365"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/2382553.2382556"},{"key":"ref29","first-page":"945","article-title":"MLaaS in the wild: Workload analysis and scheduling in large-scale heterogeneous GPU clusters","volume-title":"Proc. 19th {USENIX} Symp. Networked Syst. Des. Implementation","author":"Weng"},{"key":"ref30","first-page":"397","volume-title":"Proc. 2021 USENIX Annu. Tech. Conf.","author":"Romero"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3419111.3421285"},{"key":"ref32","first-page":"1049","article-title":"$\\lbrace${MArk $\\rbrace$}: Exploiting cloud services for $\\lbrace${ Cost-Effective $\\rbrace,\\lbrace$},{ SLO-Aware$\\rbrace$} machine learning inference serving","volume-title":"Proc. 2019 USENIX Annu. Tech. Conf.","author":"Zhang","year":"2019"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/3495243.3560551"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3299710.3211336"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/3371154"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3447993.3483249"},{"key":"ref37","article-title":"Yolov5","year":"2022"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1145\/3380985"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/n19-1423"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00881"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/3570361.3615753"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2017-614"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.14778\/3364324.3364325"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/JSAC.2023.3242724"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/3694715.3695963"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1080\/15326349.2017.1303615"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1214\/aoms\/1177728975.JSTOR2236285"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1613\/jair.3987"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1080\/23311916.2018.1502242"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1287\/opre.49.5.720.10605"},{"key":"ref51","first-page":"783","article-title":"Three steps is all you need: Fast, accurate, automatic scaling decisions for distributed streaming dataflows","volume-title":"Proc. 13th USENIX Symp. Operating Syst. Des. Implementation","author":"Kalavri"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1145\/3447993.3448628"},{"key":"ref53","volume-title":"Feedback Control Theory","author":"Doyle","year":"2013"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1145\/3230543.3230554"},{"key":"ref55","volume-title":"The Single Server Queue","author":"Cohen","year":"2012"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/WCNC.2019.8885745"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.3390\/electronics10020162"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/SC41405.2020.00073"},{"key":"ref59","first-page":"1063","article-title":"$\\lbrace${CarMap$\\rbrace$}: Fast 3D feature map updates for automobiles","volume-title":"Proc. 17th USENIX Symp. Netw. Syst. Des. Implementation","author":"Ahmad"},{"key":"ref60","article-title":"Python-for-android","year":"2022"},{"key":"ref61","first-page":"437","article-title":"A public domain dataset for human activity recognition using smartphones","volume-title":"Proc. 21th Int. Eur. Symp. Artif. Neural Netw. Computat. Intell. Mach. Learn.","author":"Anguita"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1145\/3498361.3538948"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-48421-6_6"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1038\/nature14236"},{"key":"ref66","article-title":"Load balancing 101: Nuts and bolts","volume-title":"White Paper","author":"Salchow","year":"2007"},{"key":"ref67","first-page":"11711","article-title":"Mcunet: Tiny deep learning on IoT devices","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Lin"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/SEC.2018.00016"},{"key":"ref69","first-page":"269","article-title":"Focus: Querying large video datasets with low latency and low cost","volume-title":"Proc. 13th USENIX Symp. Operating Syst. Des. Implementation","author":"Hsieh"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/IPSN.2016.7460664"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1145\/3570361.3613285"},{"key":"ref72","first-page":"1","article-title":"Convolutional neural networks with low-rank regularization","volume-title":"Proc. 4th Int. Conf. Learn. Representations","author":"Tai"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/ICC45855.2022.9839199"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1109\/TNSM.2024.3352821"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/TNSM.2024.3403714"}],"container-title":["IEEE Transactions on Mobile Computing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/7755\/11229982\/11078796.pdf?arnumber=11078796","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,6]],"date-time":"2025-11-06T18:55:16Z","timestamp":1762455316000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11078796\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12]]},"references-count":75,"journal-issue":{"issue":"12"},"URL":"https:\/\/doi.org\/10.1109\/tmc.2025.3588430","relation":{},"ISSN":["1536-1233","1558-0660","2161-9875"],"issn-type":[{"value":"1536-1233","type":"print"},{"value":"1558-0660","type":"electronic"},{"value":"2161-9875","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12]]}}}