{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,26]],"date-time":"2026-06-26T04:36:19Z","timestamp":1782448579490,"version":"3.54.5"},"reference-count":48,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"2","license":[{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE J. Emerg. Sel. Topics Circuits Syst."],"published-print":{"date-parts":[[2025,6]]},"DOI":"10.1109\/jetcas.2025.3568716","type":"journal-article","created":{"date-parts":[[2025,5,9]],"date-time":"2025-05-09T18:00:49Z","timestamp":1746813649000},"page":"217-230","source":"Crossref","is-referenced-by-count":4,"title":["End-to-End Acceleration of Generative Models With Runtime Regularized KV Cache Management"],"prefix":"10.1109","volume":"15","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-5112-300X","authenticated-orcid":false,"given":"Ashkan","family":"Moradifirouzabadi","sequence":"first","affiliation":[{"name":"Department of Electrical and Computer Engineering (ECE), University of California at San Diego, San Diego, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8104-5136","authenticated-orcid":false,"given":"Mingu","family":"Kang","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering (ECE), University of California at San Diego, San Diego, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","first-page":"114","article-title":"Keyformer: Kv cache reduction through key tokens selection for efficient generative inference","volume-title":"Proc. Mach. Learn. Syst.","volume":"6","author":"Adnan"},{"key":"ref2","first-page":"1","article-title":"Dynamic context pruning for efficient and interpretable autoregressive transformers","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Anagnostidis"},{"key":"ref3","volume-title":"Artisan Memory Compilers","year":"2024"},{"key":"ref4","article-title":"Longformer: The long-document transformer","author":"Beltagy","year":"2020","journal-title":"arXiv:2004.05150"},{"key":"ref5","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. NIPS","author":"Brown"},{"key":"ref6","volume-title":"Genus Synthesis Solution","year":"2024"},{"key":"ref7","volume-title":"Innovus Implementation System","year":"2024"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA.2016.40"},{"key":"ref9","article-title":"Generating long sequences with sparse transformers","author":"Child","year":"2019","journal-title":"arXiv:1904.10509"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA56546.2023.10071081"},{"key":"ref11","first-page":"1","article-title":"Model tells you what to discard: Adaptive KV cache compression for LLMs","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Ge"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA47549.2020.00035"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00060"},{"key":"ref14","first-page":"1","article-title":"MInference 1.0: accelerating pre-filling for long-context LLMs via dynamic sparse attention","volume-title":"Proc. 38th Annu. Conf. Neural Inf. Process. Syst.","author":"Jiang"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1145\/3079856.3080246"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3575693.3575747"},{"key":"ref17","first-page":"1","article-title":"Transformers are RNNs: Fast autoregressive transformers with linear attention","volume-title":"Proc. ICML","author":"Katharopoulos"},{"key":"ref18","article-title":"Full stack optimization of transformer inference: A survey","author":"Kim","year":"2023","journal-title":"arXiv:2302.14017"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3534678.3539260"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICCIT.2008.131"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA59077.2024.00080"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2018.2845861"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA57654.2024.00065"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3470496.3527423"},{"key":"ref25","first-page":"1","article-title":"Scissorhands: Exploiting the persistence of importance hypothesis for LLM KV cache compression at test time","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liu"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1145\/3466752.3480125"},{"key":"ref27","first-page":"1","article-title":"Pointer sentinel mixture models","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Merity"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ESSERC62670.2024.10719540"},{"key":"ref29","first-page":"37396","article-title":"Dynamic memory compression: Retrofitting LLMs for accelerated inference","volume-title":"Proc. 41st Int. Conf. Mach. Learn.","author":"Nawrot"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1145\/3620665.3640422"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3579371.3589057"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA59077.2024.00079"},{"key":"ref33","article-title":"Language models are unsupervised multitask learners","author":"Radford","year":"2019"},{"issue":"1","key":"ref34","first-page":"1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/3352460.3358302"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3665314.3670798"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1016\/j.vlsi.2017.02.002"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1145\/3466752.3480095"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA51647.2021.00018"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO61859.2024.00093"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA57654.2024.00033"},{"key":"ref43","article-title":"HuggingFace\u2019s transformers: State-of-the-art natural language processing","author":"Wolf","year":"2019","journal-title":"arXiv:1910.03771"},{"key":"ref44","first-page":"1","article-title":"Efficient streaming language models with attention sinks","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Xiao"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO56248.2022.00059"},{"key":"ref46","first-page":"521","article-title":"Orca: A distributed serving system for transformer-based generative models","volume-title":"Proc. 16th USENIX Symp. Operating Syst. Design Implement. (OSDI)","author":"Yu"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530505"},{"key":"ref48","first-page":"34661","article-title":"H2o: Heavy-hitter Oracle for efficient generative inference of large language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"36","author":"Zhang"}],"container-title":["IEEE Journal on Emerging and Selected Topics in Circuits and Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/5503868\/11050007\/10994487.pdf?arnumber=10994487","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,26]],"date-time":"2025-06-26T04:31:13Z","timestamp":1750912273000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10994487\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6]]},"references-count":48,"journal-issue":{"issue":"2"},"URL":"https:\/\/doi.org\/10.1109\/jetcas.2025.3568716","relation":{},"ISSN":["2156-3357","2156-3365"],"issn-type":[{"value":"2156-3357","type":"print"},{"value":"2156-3365","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6]]}}}