{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,21]],"date-time":"2026-04-21T14:52:26Z","timestamp":1776783146276,"version":"3.51.2"},"reference-count":14,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,5,25]],"date-time":"2025-05-25T00:00:00Z","timestamp":1748131200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,5,25]],"date-time":"2025-05-25T00:00:00Z","timestamp":1748131200000},"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":[],"published-print":{"date-parts":[[2025,5,25]]},"DOI":"10.1109\/iscas56072.2025.11043378","type":"proceedings-article","created":{"date-parts":[[2025,6,27]],"date-time":"2025-06-27T17:42:19Z","timestamp":1751046139000},"page":"1-5","source":"Crossref","is-referenced-by-count":2,"title":["Mixed INT4-INT8 LLM Quantization via Progressive Layerwise Assignment with Dynamic Sensitivity Estimation"],"prefix":"10.1109","author":[{"given":"Jihoon","family":"Kim","sequence":"first","affiliation":[{"name":"Inter-University Semiconductor Research Center (ISRC),Seoul,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Minseok","family":"Seo","sequence":"additional","affiliation":[{"name":"Inter-University Semiconductor Research Center (ISRC),Seoul,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xuan Truong","family":"Nguyen","sequence":"additional","affiliation":[{"name":"Inter-University Semiconductor Research Center (ISRC),Seoul,Korea"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3620666.3651324"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1145\/3649329.3655951"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/AICAS59952.2024.10595874"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/WACV57701.2024.00070"},{"key":"ref5","first-page":"38087","article-title":"Smoothquant: Accurate and efficient post-training quantization for large language models","volume-title":"International Conference on Machine Learning","author":"Xiao"},{"key":"ref6","first-page":"87","article-title":"Awq: Activation-aware weight quantization for on-device llm compression and acceleration","volume-title":"Proceedings of Machine Learning and Systems","volume":"6","author":"Lin"},{"key":"ref7","article-title":"Qserve: W4a8kv4 quantization and system co-design for efficient llm serving","author":"Lin","year":"2024"},{"issue":"8","key":"ref8","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI blog"},{"key":"ref9","article-title":"Opt: Open pre-trained transformer language models","author":"Zhang","year":"2022"},{"key":"ref10","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024"},{"key":"ref11","article-title":"A white paper on neural network quantization","author":"Nagel","year":"2021"},{"key":"ref12","doi-asserted-by":"crossref","DOI":"10.18653\/v1\/2021.emnlp-main.627","article-title":"Understanding and overcoming the challenges of efficient transformer quantization","author":"Bondarenko","year":"2021"},{"key":"ref13","article-title":"The pile: An 800gb dataset of diverse text for language modeling","author":"Gao","year":"2020"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P16-1144"}],"event":{"name":"2025 IEEE International Symposium on Circuits and Systems (ISCAS)","location":"London, United Kingdom","start":{"date-parts":[[2025,5,25]]},"end":{"date-parts":[[2025,5,28]]}},"container-title":["2025 IEEE International Symposium on Circuits and Systems (ISCAS)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11043142\/11042930\/11043378.pdf?arnumber=11043378","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,28]],"date-time":"2025-06-28T07:01:37Z","timestamp":1751094097000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11043378\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,25]]},"references-count":14,"URL":"https:\/\/doi.org\/10.1109\/iscas56072.2025.11043378","relation":{},"subject":[],"published":{"date-parts":[[2025,5,25]]}}}