{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T20:59:30Z","timestamp":1774645170292,"version":"3.50.1"},"reference-count":39,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"8","license":[{"start":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T00:00:00Z","timestamp":1754006400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T00:00:00Z","timestamp":1754006400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T00:00:00Z","timestamp":1754006400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100004358","name":"Samsung","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100004358","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Institute of Information and Communications Technology Planning and Evaluation"},{"name":"Korea Government (MSIT)","award":["RS 2023-00228255"],"award-info":[{"award-number":["RS 2023-00228255"]}]},{"name":"IITP"},{"name":"Korea Government (MSIT) through the Artificial Intelligence Semiconductor Support Program to Nurture the Best Talents","award":["IITP-2023-RS-2023-00256081"],"award-info":[{"award-number":["IITP-2023-RS-2023-00256081"]}]},{"name":"IITP"},{"name":"Korea Government (MSIT) through the Information Technology Research Center (ITRC) Support Program","award":["IITP-2024-2020-0-01461"],"award-info":[{"award-number":["IITP-2024-2020-0-01461"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Circuits Syst. I"],"published-print":{"date-parts":[[2025,8]]},"DOI":"10.1109\/tcsi.2024.3506999","type":"journal-article","created":{"date-parts":[[2024,12,30]],"date-time":"2024-12-30T14:48:18Z","timestamp":1735570098000},"page":"4091-4102","source":"Crossref","is-referenced-by-count":1,"title":["FACET: On-the-Fly Activation Compression for Efficient Transformer Training"],"prefix":"10.1109","volume":"72","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-5461-9234","authenticated-orcid":false,"given":"Seungyong","family":"Lee","sequence":"first","affiliation":[{"name":"Inter-University of Semiconductor Research Center (ISRC), Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Geonu","family":"Yun","sequence":"additional","affiliation":[{"name":"Inter-University of Semiconductor Research Center (ISRC), Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7527-6971","authenticated-orcid":false,"given":"Xuan","family":"Truong Nguyen","sequence":"additional","affiliation":[{"name":"Department of Next-Generation Semiconductor Convergence and Open Sharing System (COSS) and the System Semiconductor for AI Engineering (SSAI) Program, the Department of Electrical Engineering, Seoul National University, Seoul, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6811-9647","authenticated-orcid":false,"given":"Hyuk-Jae","family":"Lee","sequence":"additional","affiliation":[{"name":"Inter-University of Semiconductor Research Center (ISRC), Department of Electrical and Computer Engineering, Seoul National University, Seoul, South Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Phi-3 technical report: A highly capable language model locally on your phone","volume-title":"arXiv:2404.14219","author":"Abdin","year":"2024"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3637528.3671728"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2019.2950093"},{"key":"ref4","article-title":"Training deep nets with sublinear memory cost","author":"Chen","year":"2016","journal-title":"arXiv:1604.06174"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TVLSI.2009.2020989"},{"issue":"70","key":"ref6","first-page":"1","article-title":"Scaling instruction-finetuned language models","volume":"25","author":"Chung","year":"2024","journal-title":"J. Mach. Learn. Res."},{"key":"ref7","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv:1810.04805"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/IPDPS.2016.11"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2021.3075765"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/JRPROC.1952.273898"},{"key":"ref11","first-page":"1","article-title":"Just-in-time quantization with processing-in-memory for efficient ml training","volume-title":"Proc. 7th Annu. Conf. Mach. Learn. Syst.","author":"Ibrahim"},{"key":"ref12","article-title":"A study of BFLOAT16 for deep learning training","author":"Kalamkar","year":"2019","journal-title":"arXiv:1905.12322"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001172"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/3431815"},{"key":"ref15","first-page":"341","article-title":"Reducing activation recomputation in large transformer models","volume":"5","author":"Korthikanti","year":"2023","journal-title":"Mach. Learn. Systems"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/AICAS57966.2023.10168658"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00013"},{"key":"ref18","article-title":"BART: Denoising sequence-to-sequence pre-training for natural language generation, translation, and comprehension","author":"Lewis","year":"2019","journal-title":"arXiv:1910.13461"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2024.3350661"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/TVCG.2014.2346458"},{"key":"ref21","first-page":"14139","article-title":"Gact: Activation compressed training for generic network architectures","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Liu"},{"key":"ref22","article-title":"FP8 formats for deep learning","author":"Micikevicius","year":"2022","journal-title":"arXiv:2209.05433"},{"key":"ref23","first-page":"1","article-title":"Efficient large-scale language model training on GPU clusters using megatron-LM","volume-title":"Proc. Int. Conf. High Perform. Comput., Netw., Storage Anal.","author":"Narayanan"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2019.2962516"},{"key":"ref25","article-title":"Mesa: A memory-saving training framework for transformers","author":"Pan","year":"2021","journal-title":"arXiv:2111.11124"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/HCS52781.2021.9567119"},{"key":"ref27","first-page":"377","article-title":"Base-delta-immediate compression: Practical data compression for on-chip caches","volume-title":"Proc. 21st Int. Conf. Parallel Archit. Compilation Techn. (PACT)","author":"Pekhimenko"},{"issue":"8","key":"ref28","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"issue":"1","key":"ref29","first-page":"5485","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":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA57654.2024.00066"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2018.00017"},{"key":"ref32","article-title":"Microscaling data formats for deep learning","author":"Darvish Rouhani","year":"2023","journal-title":"arXiv:2310.10537"},{"key":"ref33","first-page":"1","article-title":"Efficient post-training quantization with fp8 formats","volume-title":"Proc. 7th Annu. Conf. Mach. Learn. Syst.","author":"Shen"},{"key":"ref34","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv:2307.09288"},{"key":"ref35","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv:2302.13971"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref37","first-page":"1","article-title":"Alam: Averaged low-precision activation for memory-efficient training of transformer models","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Woo"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2023.3337283"},{"key":"ref39","article-title":"OPT: Open pre-trained transformer language models","author":"Zhang","year":"2022","journal-title":"arXiv:2205.01068"}],"container-title":["IEEE Transactions on Circuits and Systems I: Regular Papers"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/8919\/11099062\/10817785.pdf?arnumber=10817785","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,27]],"date-time":"2026-03-27T19:54:13Z","timestamp":1774641253000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10817785\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8]]},"references-count":39,"journal-issue":{"issue":"8"},"URL":"https:\/\/doi.org\/10.1109\/tcsi.2024.3506999","relation":{},"ISSN":["1549-8328","1558-0806"],"issn-type":[{"value":"1549-8328","type":"print"},{"value":"1558-0806","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8]]}}}