{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T13:23:07Z","timestamp":1773840187862,"version":"3.50.1"},"reference-count":42,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"7","license":[{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key Research and Development Program of China","doi-asserted-by":"publisher","award":["2022YFB4501000"],"award-info":[{"award-number":["2022YFB4501000"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62202328"],"award-info":[{"award-number":["62202328"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Comput."],"published-print":{"date-parts":[[2025,7]]},"DOI":"10.1109\/tc.2025.3558009","type":"journal-article","created":{"date-parts":[[2025,4,8]],"date-time":"2025-04-08T17:25:31Z","timestamp":1744133131000},"page":"2195-2209","source":"Crossref","is-referenced-by-count":5,"title":["TightLLM: Maximizing Throughput for LLM Inference via Adaptive Offloading Policy"],"prefix":"10.1109","volume":"74","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-0458-0900","authenticated-orcid":false,"given":"Yitao","family":"Hu","sequence":"first","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4746-5599","authenticated-orcid":false,"given":"Xiulong","family":"Liu","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-9355-9296","authenticated-orcid":false,"given":"Guotao","family":"Yang","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-7315-950X","authenticated-orcid":false,"given":"Linxuan","family":"Li","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0001-6488-3813","authenticated-orcid":false,"given":"Kai","family":"Zeng","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-7865-4905","authenticated-orcid":false,"given":"Zhixin","family":"Zhao","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7038-4407","authenticated-orcid":false,"given":"Sheng","family":"Chen","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1967-2192","authenticated-orcid":false,"given":"Laiping","family":"Zhao","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8507-0339","authenticated-orcid":false,"given":"Wenxin","family":"Li","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1758-3030","authenticated-orcid":false,"given":"Keqiu","family":"Li","sequence":"additional","affiliation":[{"name":"Tianjin Key Laboratory of Advanced Networking, Department of Intelligence and Computing, Tianjin University, Tianjin, China"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3512467"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3560260"},{"key":"ref3","article-title":"A paradigm shift in machine translation: Boosting translation performance of large language models","volume-title":"Proc. ICLR","author":"Xu","year":"2024"},{"key":"ref4","first-page":"41092","article-title":"Prompting large language model for machine translation: A case study","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Zhang","year":"2023"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2020.3022318"},{"key":"ref6","first-page":"1661","article-title":"Spreadsheetcoder: Formula prediction from semi-structured context","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Chen","year":"2021"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1016\/j.hcc.2024.100211"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CNS62487.2024.10735498"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2023.3305754"},{"key":"ref12","article-title":"GPT-4 technical report","author":"Achiam","year":"2023"},{"key":"ref13","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023"},{"key":"ref14","article-title":"Fastertransformer","year":"2024"},{"key":"ref15","first-page":"559","article-title":"Alpa: Automating inter-and $\\{${intra-operator$\\}$} parallelism for distributed deep learning","volume-title":"Proc. OSDI","author":"Zheng","year":"2022"},{"key":"ref16","first-page":"663","article-title":"$\\{${AlpaServe$\\}$}: Statistical multiplexing with model parallelism for deep learning serving","volume-title":"Proc. OSDI","author":"Li","year":"2023"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3694715.3695948"},{"key":"ref18","first-page":"30318","article-title":"GPT3. int8 () 8-bit matrix multiplication for transformers at scale","volume":"35","author":"Dettmers","year":"2022","journal-title":"Proc. Adv. Neural Inf. Process. Syst."},{"key":"ref19","first-page":"27168","article-title":"ZeroQuant: Efficient and affordable post-training quantization for large-scale transformers","volume":"35","author":"Yao","year":"2022","journal-title":"Proc. Adv. Neural Inf. Process. Syst."},{"key":"ref20","article-title":"Model tells you what to discard: Adaptive KV cache compression for LLMs","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Ge","year":"2024"},{"key":"ref21","article-title":"Efficient streaming language models with attention sinks","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Xiao","year":"2024"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2020.2995593"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/3394486.3406703"},{"key":"ref24","article-title":"Accelerate: Training and inference at scale made simple, efficient and adaptable","author":"Gugger","year":"2022"},{"key":"ref25","first-page":"31094","article-title":"FlexGen: High-throughput generative inference of large language models with a single GPU","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Sheng","year":"2023"},{"key":"ref26","first-page":"162","article-title":"HeteGen: Efficient heterogeneous parallel inference for large language models on resource-constrained devices","volume":"6","author":"Zhao","year":"2024","journal-title":"Proc. MLSys"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2020.3027671"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2024.3365939"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2024.3404051"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2022.3208206"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2024.3389507"},{"key":"ref33","article-title":"Scaling laws for neural language models","author":"Kaplan","year":"2020"},{"key":"ref34","article-title":"OPT: Open pre-trained transformer language models","author":"Zhang","year":"2022"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/3656019.3676949"},{"key":"ref37","first-page":"155","article-title":"$\\{${InfiniGen$\\}$}: Efficient generative inference of large language models with dynamic $\\{${KV$\\}$} cache management","volume-title":"Proc. OSDI","author":"Lee","year":"2024"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.5555\/3454287.3455008"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1145\/3694715.3695964"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.678"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2021.3125792"},{"key":"ref42","first-page":"62557","article-title":"SGLang: Efficient execution of structured language model programs","volume-title":"Proc. 38th Annu. Conf. Neural Inf. Process. Syst.","author":"Zheng","year":"2024"},{"key":"ref43","first-page":"111","article-title":"$\\{${Cost-Efficient$\\}$} large language model serving for multi-turn conversations with $\\{${CachedAttention$\\}$}","volume-title":"Proc. USENIX Annu. Tech. Conf. (USENIX ATC)","author":"Gao","year":"2024"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1145\/3689031.3696072"},{"key":"ref45","first-page":"16344","article-title":"FlashAttention: Fast and memory-efficient exact attention with IO-awareness","volume":"35","author":"Dao","year":"2022","journal-title":"Proc. Adv. Neural Inf. Process. Syst."}],"container-title":["IEEE Transactions on Computers"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/12\/11030070\/10949701.pdf?arnumber=10949701","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,12]],"date-time":"2025-06-12T19:24:38Z","timestamp":1749756278000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10949701\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7]]},"references-count":42,"journal-issue":{"issue":"7"},"URL":"https:\/\/doi.org\/10.1109\/tc.2025.3558009","relation":{},"ISSN":["0018-9340","1557-9956","2326-3814"],"issn-type":[{"value":"0018-9340","type":"print"},{"value":"1557-9956","type":"electronic"},{"value":"2326-3814","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7]]}}}