{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,13]],"date-time":"2026-10-13T18:35:58Z","timestamp":1791916558439,"version":"4.3.4"},"reference-count":45,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"10","license":[{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100014219","name":"National Natural Science Foundation of China Distinguished Young Scholars Program","doi-asserted-by":"publisher","award":["62125208"],"award-info":[{"award-number":["62125208"]}],"id":[{"id":"10.13039\/501100014219","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62325405"],"award-info":[{"award-number":["62325405"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U19B2019"],"award-info":[{"award-number":["U19B2019"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U21B2031"],"award-info":[{"award-number":["U21B2031"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["92364201"],"award-info":[{"award-number":["92364201"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62222102"],"award-info":[{"award-number":["62222102"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62171023"],"award-info":[{"award-number":["62171023"]}],"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":[[2025,10]]},"DOI":"10.1109\/tc.2025.3587957","type":"journal-article","created":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T17:56:55Z","timestamp":1752775015000},"page":"3390-3403","source":"Crossref","is-referenced-by-count":1,"title":["WOLF: <u>W<\/u>eight-Level <u>O<\/u>ut<u>L<\/u>ier and <u>F<\/u>ault Integration for Reliable LLM Deployment"],"prefix":"10.1109","volume":"74","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-8837-6815","authenticated-orcid":false,"given":"Chong","family":"Wang","sequence":"first","affiliation":[{"name":"Department of Electronic Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9851-1970","authenticated-orcid":false,"given":"Wanyi","family":"Fu","sequence":"additional","affiliation":[{"name":"Department of Electronic Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2155-6318","authenticated-orcid":false,"given":"Jiangwei","family":"Zhang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Cyberspace Security Defense, Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-4755-6881","authenticated-orcid":false,"given":"Shiyao","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Electronic Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9215-7632","authenticated-orcid":false,"given":"Rui","family":"Hou","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Cyberspace Security Defense, Institute of Information Engineering, Chinese Academy of Sciences, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0036-9233","authenticated-orcid":false,"given":"Jian","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Electronic Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6108-5157","authenticated-orcid":false,"given":"Yu","family":"Wang","sequence":"additional","affiliation":[{"name":"Department of Electronic Engineering, Tsinghua University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"A survey of large language models","author":"Zhao","year":"2023"},{"key":"ref2","article-title":"OPT: Open pre-trained transformer language models","author":"Zhang","year":"2022"},{"key":"ref3","first-page":"38087","article-title":"SmoothQuant: Accurate and efficient post-training quantization for large language models","volume-title":"Proc. ICML","author":"Xiao","year":"2023"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.23919\/DATE48585.2020.9116571"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TPDS.2020.3043449"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA.2018.00015"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA47549.2020.00037"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO50266.2020.00071"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3579371.3589038"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ISVLSI54635.2022.00065"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1145\/1816038.1815980"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1145\/2508148.2485929"},{"key":"ref13","article-title":"GPTQ: Accurate post-training quantization for generative pre-trained transformers","author":"Frantar","year":"2022"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1145\/3714983.3714987"},{"key":"ref15","article-title":"Hello gpt-4o"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1810.04805"},{"key":"ref17","article-title":"Llama 3"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/362686.362692"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO56248.2022.00095"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1145\/3007787.3001165"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TNANO.2013.2253329"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/IEDM.2007.4418973"},{"key":"ref23","first-page":"32","article-title":"Pytorch: An imperative style, high-performance deep learning library","volume-title":"Proc. NIPS","author":"Paszke","year":"2019"},{"issue":"8","key":"ref24","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"key":"ref25","article-title":"Bloom: A 176b-parameter open-access multilingual language model","author":"Le Scao","year":"2022"},{"key":"ref26","article-title":"LLAMA 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6239"},{"key":"ref28","article-title":"Think you have solved question answering? Try Arc, the AI2 reasoning challenge","author":"Clark","year":"2018"},{"key":"ref29","article-title":"ModelTC"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA56546.2023.10071093"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1002\/j.1538-7305.1950.tb00463.x"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2012.2185930"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/MM.2019.2892165"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1201\/9781003162810-13"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1145\/3394885.3431554"},{"key":"ref36","first-page":"7543","article-title":"Improving neural network quantization without retraining using outlier channel splitting","volume-title":"Proc. ICML","author":"Zhao","year":"2019"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1145\/3289602.3293964"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP39728.2021.9414076"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i9.16950"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1145\/3316781.3317783"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA57654.2024.00082"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/MDAT.2023.3241116"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/JETCAS.2017.2776980"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/ICCD46524.2019.00020"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18074.2021.9586269"}],"container-title":["IEEE Transactions on Computers"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/12\/11164542\/11082756.pdf?arnumber=11082756","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,16]],"date-time":"2025-09-16T17:33:46Z","timestamp":1758044026000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11082756\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10]]},"references-count":45,"journal-issue":{"issue":"10"},"URL":"https:\/\/doi.org\/10.1109\/tc.2025.3587957","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":[[2025,10]]}}}