{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T05:01:10Z","timestamp":1780117270953,"version":"3.54.0"},"reference-count":94,"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:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,7,1]],"date-time":"2026-07-01T00:00:00Z","timestamp":1782864000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Science and Technology Development Fund of Macau","award":["0107\/2024\/RIA2"],"award-info":[{"award-number":["0107\/2024\/RIA2"]}]},{"name":"Science and Technology Development Fund of Macau","award":["0061\/2025\/RIB2"],"award-info":[{"award-number":["0061\/2025\/RIB2"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Parallel Distrib. Syst."],"published-print":{"date-parts":[[2026,7]]},"DOI":"10.1109\/tpds.2026.3665358","type":"journal-article","created":{"date-parts":[[2026,2,16]],"date-time":"2026-02-16T21:09:00Z","timestamp":1771276140000},"page":"1630-1646","source":"Crossref","is-referenced-by-count":1,"title":["<b>Floe<\/b>\n                    : Federated Specialization for Real-Time LLM\u2013SLM Inference"],"prefix":"10.1109","volume":"37","author":[{"ORCID":"https:\/\/orcid.org\/0009-0009-5220-1609","authenticated-orcid":false,"given":"Chunlin","family":"Tian","sequence":"first","affiliation":[{"name":"State Key Laboratory of Internet of Things for Smart City, University of Macau, Macau, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Kahou","family":"Tam","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Internet of Things for Smart City, University of Macau, Macau, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2422-1356","authenticated-orcid":false,"given":"Yebo","family":"Wu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Internet of Things for Smart City, University of Macau, Macau, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-4574-0454","authenticated-orcid":false,"given":"Shuaihang","family":"Zhong","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Internet of Things for Smart City, University of Macau, Macau, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2044-8289","authenticated-orcid":false,"given":"Li","family":"Li","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Internet of Things for Smart City, University of Macau, Macau, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2728-8273","authenticated-orcid":false,"given":"Nicholas D.","family":"Lane","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, University of Cambridge, England, U.K."}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9480-0356","authenticated-orcid":false,"given":"ChengZhong","family":"Xu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Internet of Things for Smart City, University of Macau, Macau, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"GPT-4 technical report","year":"2023"},{"key":"ref2","first-page":"1","article-title":"PALM: Scaling language modeling with pathways","volume":"24","author":"Chowdhery","year":"2023","journal-title":"J. Mach. Learn. Res."},{"key":"ref3","article-title":"Qwen technical report","author":"Bai","year":"2023"},{"key":"ref4","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023"},{"key":"ref5","article-title":"Open models based on gemini research and technology","author":"Mesnard","year":"2024"},{"key":"ref6","article-title":"Language models are few-shot learners","author":"Brown","year":"2020"},{"key":"ref7","article-title":"GitHub Copilot: Your AI pair programmer"},{"key":"ref8","article-title":"Your everyday AI companion| Microsoft Bing"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.7759\/cureus.40895"},{"key":"ref10","article-title":"ChatLaw: Open-source legal large language model with integrated external knowledge bases","author":"Cui","year":"2023"},{"key":"ref11","article-title":"BloombergGPT: A large language model for finance","author":"Wu","year":"2023"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/3637528.3671573"},{"key":"ref13","article-title":"Federated learning: Strategies for improving communication efficiency","author":"Kone\u010dn\u00fd","year":"2016"},{"key":"ref14","article-title":"AI Index: State of AI in 13 charts","year":"2024","journal-title":"Stanford Inst. Hum.-Centered Artif. Intell."},{"key":"ref15","article-title":"DeepSeek-V3 technical report","year":"2024"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.445"},{"key":"ref17","article-title":"When federated learning meets pre-trained language models\u2019 parameter-efficient tuning methods","author":"Zhang","year":"2022"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3637528.3671573"},{"key":"ref19","article-title":"FATE-LLM: A industrial grade federated learning framework for large language models","author":"Fan","year":"2023"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/icassp48485.2024.10447454"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP49357.2023.10095356"},{"key":"ref22","article-title":"Offsite-tuning: Transfer learning without full model","author":"Xiao","year":"2023"},{"key":"ref23","first-page":"27168","article-title":"ZeroQuant: Efficient and affordable post-training quantization for large-scale transformers","volume-title":"Proc. Adv. Neural Inf. Process. Syst. 35: Annu. Conf. Neural Inf. Process. Syst.","author":"Yao"},{"key":"ref24","article-title":"FediPR: Ownership verification for federated deep neural network models","author":"Fan","year":"2021"},{"key":"ref25","first-page":"21702","article-title":"LLM-pruner: On the structural pruning of large language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ma"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/tdsc.2024.3390761"},{"key":"ref27","first-page":"1057","article-title":"FLASH: Towards a high-performance hardware acceleration architecture for cross-silo federated learning","volume-title":"Proc. 20th USENIX Symp. Netw. Syst. Des. Implementation","author":"Zhang"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.14722\/ndss.2021.24119"},{"key":"ref29","article-title":"Gemini: A family of highly capable multimodal models","author":"Team","year":"2023"},{"key":"ref30","first-page":"1","article-title":"LoRA: Low-rank adaptation of large language models","volume-title":"Proc. 10th Int. Conf. Learn. Representations","author":"Hu"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1991.3.1.79"},{"key":"ref32","article-title":"Jetson Nano","year":"2023"},{"key":"ref33","article-title":"Will we run out of data? an analysis of the limits of scaling datasets in machine learning","author":"Villalobos","year":"2022"},{"key":"ref34","article-title":"Using logit bias to alter token probability with the openai API","year":"2024"},{"key":"ref35","article-title":"Tuning language models by proxy","author":"Liu","year":"2024"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D19-5606"},{"key":"ref37","article-title":"CPT: Consistent proxy tuning for black-box optimization","author":"He","year":"2024"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.235"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.754"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.687"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01544"},{"key":"ref42","article-title":"LLM.int8(): 8-bit matrix multiplication for transformers at scale","author":"Dettmers","year":"2022"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.249"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.findings-emnlp.976"},{"key":"ref45","first-page":"2924","article-title":"Boolq exploring the surprising difficulty of natural yes\/no questions","volume-title":"Proc. Conf. North Amer. Chapter Assoc. Comput. Linguistics: Hum. Lang. Technol.","author":"Clark"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/p19-1472"},{"key":"ref47","first-page":"2381","volume-title":"Proc. 2018 Conf. Empirical Methods Natural Lang. Process.","author":"Mihaylov"},{"key":"ref48","first-page":"21702","article-title":"LLM-pruner: On the structural pruning of large language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst. 36: Annu. Conf. Neural Inf. Process. Syst.","author":"Ma"},{"key":"ref49","article-title":"Splitnn-driven vertical partitioning","author":"Ceballos","year":"2020"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2025.3558009"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2024.3480115"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.206"},{"key":"ref53","article-title":"Measuring mathematical problem solving with the math dataset","author":"Hendrycks","year":"2021"},{"key":"ref54","article-title":"Beyond the imitation game: Quantifying and extrapolating the capabilities of language models","author":"Srivastava","year":"2022"},{"key":"ref55","first-page":"1","article-title":"Measuring massive multitask language understanding","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Hendrycks"},{"key":"ref56","article-title":"Open llm leaderboard V2","author":"Fourrier","year":"2024"},{"key":"ref57","article-title":"Qwen2 technical report","author":"Team","year":"2024"},{"key":"ref58","article-title":"The LLAMA 3 herd of models","author":"Grattafiori","year":"2024"},{"key":"ref59","article-title":"Federated learning with non-IID data","author":"Zhao","year":"2018"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO56248.2022.00049"},{"key":"ref61","article-title":"On the convergence of federated optimization in heterogeneous networks","author":"Sahu","year":"2018"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/INFOCOM41043.2020.9155494"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52688.2022.00982"},{"key":"ref64","article-title":"Code Alpaca: An instruction-following llama model for code generation","author":"Chaudhary","year":"2023"},{"key":"ref65","article-title":"Evaluating large language models trained on code","author":"Chen","year":"2021"},{"key":"ref66","article-title":"BGE M3-embedding: Multi-lingual, multi-functionality, multi-granularity text embeddings through self-knowledge distillation","author":"Chen","year":"2024","journal-title":"arXiv:2402.03216"},{"key":"ref67","article-title":"Finetuned language models are zero-shot learners","author":"Wei","year":"2021"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2015.7056028"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/ICDCS47774.2020.00046"},{"key":"ref70","first-page":"19","article-title":"OORT: Efficient federated learning via guided participant selection","volume-title":"Proc. 15th USENIX Symp. Operating Syst. Des. Implementation","author":"Lai"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1145\/3605573.3605584"},{"key":"ref72","article-title":"ZeroFL: Efficient on-device training for federated learning with local sparsity","author":"Qiu","year":"2022"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02601"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1145\/2976749.2978318"},{"key":"ref76","article-title":"Differentially private federated learning: A client level perspective","author":"Geyer","year":"2017"},{"key":"ref77","article-title":"Learning differentially private recurrent language models","author":"McMahan","year":"2017"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/TIFS.2020.2988575"},{"key":"ref79","article-title":"GShard: Scaling giant models with conditional computation and automatic sharding","author":"Lepikhin","year":"2020"},{"key":"ref80","article-title":"Switch Transformers: Scaling to trillion parameter models with simple and efficient sparsity","author":"Fedus","year":"2022"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.5555\/3454287.3455008"},{"key":"ref82","article-title":"Capture browser trace in the Azure portal","year":"2024"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/ICFEC54809.2022.00012"},{"key":"ref84","article-title":"Benchmarking in the dark: On the absence of comprehensive edge datasets","volume-title":"Proc. 3rd USENIX Workshop Hot Topics Edge Comput.","author":"Kolosov"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1109\/ISPASS.2018.00015"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/IISWC.2013.6704679"},{"key":"ref87","article-title":"Track and reduce CO2 emissions from your computing","year":"2023"},{"key":"ref88","article-title":"TinyLlama: An open-source small language model","author":"Zhang","year":"2024"},{"key":"ref89","article-title":"Finetuned language models are zero-shot learners","volume-title":"Proc. 10th Int. Conf. Learn. Representations","author":"Wei"},{"key":"ref90","article-title":"Pointer sentinel mixture models","volume-title":"Proc. 5th Int. Conf. Learn. Representations","author":"Merity"},{"key":"ref91","article-title":"Federated learning of deep networks using model averaging","author":"McMahan","year":"2016"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i8.20819"},{"key":"ref93","first-page":"25401","article-title":"Pushing mixture of experts to the limit: Extremely parameter efficient MoE for instruction tuning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zadouri","year":"2024"},{"key":"ref94","first-page":"1","article-title":"Stanford Alpaca: An instruction-following llama model","author":"Taori","year":"2023"}],"container-title":["IEEE Transactions on Parallel and Distributed Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/71\/11520459\/11397454.pdf?arnumber=11397454","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,30]],"date-time":"2026-05-30T04:16:40Z","timestamp":1780114600000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11397454\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,7]]},"references-count":94,"journal-issue":{"issue":"7"},"URL":"https:\/\/doi.org\/10.1109\/tpds.2026.3665358","relation":{},"ISSN":["1045-9219","1558-2183","2161-9883"],"issn-type":[{"value":"1045-9219","type":"print"},{"value":"1558-2183","type":"electronic"},{"value":"2161-9883","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,7]]}}}