{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,11]],"date-time":"2026-03-11T01:51:43Z","timestamp":1773193903550,"version":"3.50.1"},"reference-count":42,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Seatrium New Energy Laboratory, Singapore Ministry of Education (MOE) Tier 1","award":["RT5\/23"],"award-info":[{"award-number":["RT5\/23"]}]},{"name":"Seatrium New Energy Laboratory, Singapore Ministry of Education (MOE) Tier 1","award":["RG24\/24"],"award-info":[{"award-number":["RG24\/24"]}]},{"name":"Nanyang Technological University (NTU) Centre for Computational Technologies in Finance"},{"name":"Research Innovation and Enterprise (RIE) 2025 Industry Alignment Fund-Industry Collaboration Projects (IAF-ICP), administered by the Agency for Science, Technology and Research","award":["I2301E0026"],"award-info":[{"award-number":["I2301E0026"]}]},{"name":"National Science Foundation","award":["CPS-2313109"],"award-info":[{"award-number":["CPS-2313109"]}]},{"DOI":"10.13039\/100000006","name":"Office of Naval Research","doi-asserted-by":"crossref","award":["N00014-22-1-2305"],"award-info":[{"award-number":["N00014-22-1-2305"]}],"id":[{"id":"10.13039\/100000006","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/100000181","name":"Air Force Office of Scientific Research","doi-asserted-by":"crossref","award":["FA9550-24-1-0083"],"award-info":[{"award-number":["FA9550-24-1-0083"]}],"id":[{"id":"10.13039\/100000181","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Netw."],"published-print":{"date-parts":[[2026]]},"DOI":"10.1109\/ton.2026.3669011","type":"journal-article","created":{"date-parts":[[2026,2,27]],"date-time":"2026-02-27T20:48:33Z","timestamp":1772225313000},"page":"3808-3823","source":"Crossref","is-referenced-by-count":1,"title":["Serving Long-Context LLMs at the Mobile Edge: Test-Time Reinforcement Learning-Based Model Caching and Inference Offloading"],"prefix":"10.1109","volume":"34","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-8203-8146","authenticated-orcid":false,"given":"Minrui","family":"Xu","sequence":"first","affiliation":[{"name":"College of Computing and Data Science, Nanyang Technological University, Nanyang Avenue, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7442-7416","authenticated-orcid":false,"given":"Dusit","family":"Niyato","sequence":"additional","affiliation":[{"name":"College of Computing and Data Science, Nanyang Technological University, Nanyang Avenue, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2771-3521","authenticated-orcid":false,"given":"Christopher G.","family":"Brinton","sequence":"additional","affiliation":[{"name":"Elmore Family School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN, USA"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/MWC.005.2400019"},{"key":"ref2","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. NIPS","author":"Brown"},{"key":"ref3","article-title":"Sparks of artificial general intelligence: Early experiments with GPT-4","author":"Bubeck","year":"2023","journal-title":"arXiv:2303.12712"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/COMST.2024.3465447"},{"key":"ref5","article-title":"The CAP principle for LLM serving","author":"Zeng","year":"2024","journal-title":"arXiv:2405.11299"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM52122.2024.10621342"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM48880.2022.9796705"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1145\/3565287.3610264"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2022.3189186"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155455"},{"key":"ref11","article-title":"MemServe: Context caching for disaggregated LLM serving with elastic memory pool","author":"Hu","year":"2024","journal-title":"arXiv:2406.17565"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICECS61496.2024.10848749"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1800"},{"key":"ref14","article-title":"Self-consistency improves chain of thought reasoning in language models","author":"Wang","year":"2022","journal-title":"arXiv:2203.11171"},{"key":"ref15","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017","journal-title":"arXiv:1707.06347"},{"key":"ref16","article-title":"Learning to (learn at test time): RNNs with expressive hidden states","author":"Sun","year":"2024","journal-title":"arXiv:2407.04620"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/MCOM.001.2400764"},{"key":"ref18","article-title":"PerLLM: Personalized inference scheduling with edge-cloud collaboration for diverse LLM services","author":"Yang","year":"2024","journal-title":"arXiv:2405.14636"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM52122.2024.10621164"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/IWCMC61514.2024.10592339"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM48880.2022.9796714"},{"key":"ref22","article-title":"TrimCaching: Parameter-sharing edge caching for AI model downloading","author":"Qu","year":"2024","journal-title":"arXiv:2404.14204"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOMWKSHPS54753.2022.9798315"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3229543.3229555"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1016\/j.comnet.2024.110564"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICC42927.2021.9500929"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/JSAC.2023.3310080"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2024.3415661"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.2307\/2234848"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/ICC45855.2022.9838736"},{"key":"ref31","first-page":"1","article-title":"Are transformers universal approximators of sequence-to-sequence functions?","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Yun"},{"key":"ref32","article-title":"Why can large language models generate correct chain-of-thoughts?","author":"Tutunov","year":"2023","journal-title":"arXiv:2310.13571"},{"key":"ref33","article-title":"A latent space theory for emergent abilities in large language models","author":"Jiang","year":"2023","journal-title":"arXiv:2304.09960"},{"key":"ref34","article-title":"Cached model-as-a-resource: Provisioning large language model agents for edge intelligence in space-air-ground integrated networks","author":"Xu","year":"2024","journal-title":"arXiv:2403.05826"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01457"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/d15-1202"},{"key":"ref37","article-title":"Think you have solved question answering? try ARC, the AI2 reasoning challenge","author":"Clark","year":"2018","journal-title":"arXiv:1803.05457"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00370"},{"key":"ref39","article-title":"CommonsenseQA: A question answering challenge targeting commonsense knowledge","author":"Talmor","year":"2018","journal-title":"arXiv:1811.00937"},{"key":"ref40","article-title":"Adversarial GLUE: A multi-task benchmark for robustness evaluation of language models","author":"Wang","year":"2021","journal-title":"arXiv:2111.02840"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP.2019.8683090"},{"key":"ref42","article-title":"Training verifiers to solve math word problems","author":"Cobbe","year":"2021","journal-title":"arXiv:2110.14168"}],"container-title":["IEEE Transactions on Networking"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/10723154\/11317935\/11417298-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/10723154\/11317935\/11417298.pdf?arnumber=11417298","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T05:23:28Z","timestamp":1773120208000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11417298\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":42,"URL":"https:\/\/doi.org\/10.1109\/ton.2026.3669011","relation":{},"ISSN":["2998-4157"],"issn-type":[{"value":"2998-4157","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026]]}}}