{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,7]],"date-time":"2026-10-07T17:09:47Z","timestamp":1791392987975,"version":"4.3.3"},"reference-count":89,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"7","license":[{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"Research Grants Council (RGC) of Hong Kong"},{"name":"General Research Fund","award":["14207123"],"award-info":[{"award-number":["14207123"]}]},{"name":"Strategic Topics","award":["STG1\/E-403\/24-N"],"award-info":[{"award-number":["STG1\/E-403\/24-N"]}]},{"name":"Strategic Topics","award":["T43-513\/23-N"],"award-info":[{"award-number":["T43-513\/23-N"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. on Mobile Comput."],"published-print":{"date-parts":[[2026,7]]},"DOI":"10.1109\/tmc.2026.3666311","type":"journal-article","created":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T20:46:49Z","timestamp":1772138809000},"page":"10740-10756","source":"Crossref","is-referenced-by-count":2,"title":["An Efficient Edge-Cloud Collaboration System With Foundational Models for Open-Set IoT Applications"],"prefix":"10.1109","volume":"25","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-0032-2539","authenticated-orcid":false,"given":"Bufang","family":"Yang","sequence":"first","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Wenrui","family":"Lu","sequence":"additional","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2130-1385","authenticated-orcid":false,"given":"Lixing","family":"He","sequence":"additional","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2072-1502","authenticated-orcid":false,"given":"Neiwen","family":"Ling","sequence":"additional","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4433-5211","authenticated-orcid":false,"given":"Zhenyu","family":"Yan","sequence":"additional","affiliation":[{"name":"Department of Information Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, 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, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xian","family":"Shuai","sequence":"additional","affiliation":[{"name":"Noah’s Ark Lab in Huawei Technologies, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiaozhe","family":"Ren","sequence":"additional","affiliation":[{"name":"Noah’s Ark Lab in Huawei Technologies, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xin","family":"Jiang","sequence":"additional","affiliation":[{"name":"Noah’s Ark Lab in Huawei Technologies, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Revenues from the AI-driven hardware market worldwide from 2018 to 2025"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3495243.3560519"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TIM.2023.3346527"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1145\/3636534.3649352"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/3666025.3699428"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TIM.2022.3154838"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3302506.3310391"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3356250.3360020"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/WACV48630.2021.00156"},{"key":"ref10","article-title":"Gpt-4 Technical Report","author":"Achiam","year":"2023"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01457"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3699765"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3712286"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/FMSys62467.2024.00010"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3625687.3625793"},{"key":"ref16","article-title":"On the opportunities and risks of foundation models","author":"Bommasani","year":"2021"},{"key":"ref17","article-title":"FreeLM: Fine-tuning-free language model","author":"Li","year":"2023"},{"key":"ref18","article-title":"Azure openAI service","year":"2024"},{"key":"ref19","first-page":"492","article-title":"LM-NAV: Robotic navigation with large pre-trained models of language, vision, and action","volume-title":"Conf. Robot Learn.","author":"Shah","year":"2023"},{"key":"ref20","article-title":"Do as i can and not as i say: Grounding language in robotic affordances","author":"Ahn","year":"2022"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01423"},{"key":"ref22","article-title":"Contrastive learning rivals masked image modeling in fine-tuning via feature distillation","author":"Wei","year":"2022"},{"key":"ref23","article-title":"Faster segment anything: Towards lightweight sam for mobile applications","author":"Zhang","year":"2023"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3093337.3037698"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/3372224.3419194"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1145\/3450268.3453520"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref28","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Radford","year":"2021"},{"key":"ref29","article-title":"LanguageBind: Extending video-language pretraining to N-modality by language-based semantic alignment","volume-title":"Proc. 12th Int. Conf. Learn. Representations","author":"Zhu","year":"2024"},{"key":"ref30","first-page":"34892","article-title":"Visual instruction tuning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"36","author":"Liu","year":"2024"},{"key":"ref31","article-title":"Qwen-vl: A versatile vision-language model for understanding, localization, text reading, and beyond","author":"Bai","year":"2023"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.3233\/faia240942"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2024.3504859"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2025.3546466"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/JIOT.2024.3524255"},{"key":"ref36","article-title":"PICE: A semantic-driven progressive inference system for LLM serving in cloud-edge networks","author":"Zhan","year":"2025"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00773"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.01338"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3302506.3310402"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58542-6_29"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.521"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1145\/3534619"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/3495243.3517016"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1145\/3241539.3241563"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1145\/3373376.3378534"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1145\/3552326.3587438"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.257"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.681"},{"key":"ref50","article-title":"Gshard: Scaling giant models with conditional computation and automatic sharding","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Lepikhin","year":"2021"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1145\/3495243.3560551"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1145\/3384419.3430898"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/ICVGIP.2008.47"},{"key":"ref55","article-title":"Mobilenets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/TII.2019.2912809"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.findings-emnlp.372"},{"key":"ref59","first-page":"917","article-title":"Recl: Responsive resource-efficient continuous learning for video analytics","volume-title":"Proc. 20th USENIX Symp. Networked Syst. Des. Implementation","author":"Khani","year":"2023"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-acl.490"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2017.2773081"},{"key":"ref62","article-title":"Experience replay for continual learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Rolnick","year":"2019"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1016\/j.patrec.2009.09.011"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18074.2021.9586176"},{"key":"ref65","first-page":"1321","article-title":"On calibration of modern neural networks","volume-title":"Int. Conf. Mach. Learn.","author":"Guo","year":"2017"},{"key":"ref66","article-title":"UCF101: A dataset of 101 human actions classes from videos in the wild","author":"Soomro","year":"2012"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/CODESISSS.2015.7331375"},{"key":"ref68","first-page":"119","article-title":"Ekya: Continuous learning of video analytics models on edge compute servers","volume-title":"19th USENIX Symp. Networked Syst. Des. Implementation","author":"Bhardwaj","year":"2022"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1145\/381677.381705"},{"key":"ref70","article-title":"iPERF","author":"Jon","year":"2014"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1145\/3446382.3448359"},{"key":"ref72","article-title":"A closer look at few-shot classification","author":"Chen","year":"2019"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2021.3065234"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1310.4546"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1810.04805"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1145\/2733373.2806390"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-33715-4_54"},{"key":"ref78","article-title":"Free teledyne flir thermal dataset for algorithm training","year":"2015"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298655"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2017.2723009"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1145\/3387514.3405874"},{"key":"ref82","article-title":"Distilling the knowledge in a neural network","volume-title":"Proc. NIPS Deep Learn. Representation Learn. Workshop","author":"Hinton","year":"2015"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19833-5_29"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01460"},{"key":"ref85","article-title":"ContextAgent: Context-aware proactive LLM agents with open-world sensory perceptions","volume-title":"Proc. 39th Annu. Conf. Neural Inf. Process. Syst.","author":"Yang","year":"2025"},{"key":"ref86","article-title":"ProAgent: Harnessing on-demand sensory contexts for proactive LLM agent systems","author":"Yang","year":"2025"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1145\/3715014.3722070"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20083-0_9"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-05318-5_3"},{"key":"ref90","article-title":"LoRA: Low-rank adaptation of large language models","volume-title":"Int. Conf. Learn. Representations","author":"Hu","year":"2021"}],"container-title":["IEEE Transactions on Mobile Computing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/7755\/11552380\/11415373.pdf?arnumber=11415373","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,6,8]],"date-time":"2026-06-08T19:55:02Z","timestamp":1780948502000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11415373\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7]]},"references-count":89,"journal-issue":{"issue":"7"},"URL":"https:\/\/doi.org\/10.1109\/tmc.2026.3666311","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":[[2026,7]]}}}