{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T17:03:35Z","timestamp":1770743015926,"version":"3.49.0"},"reference-count":37,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,6,22]],"date-time":"2025-06-22T00:00:00Z","timestamp":1750550400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,6,22]],"date-time":"2025-06-22T00:00:00Z","timestamp":1750550400000},"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","id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,6,22]]},"DOI":"10.1109\/dac63849.2025.11132816","type":"proceedings-article","created":{"date-parts":[[2025,9,15]],"date-time":"2025-09-15T17:35:41Z","timestamp":1757957741000},"page":"1-7","source":"Crossref","is-referenced-by-count":1,"title":["DuoQ: A DSP Utilization-aware and Outlier-free Quantization for FPGA-based LLMs Acceleration"],"prefix":"10.1109","author":[{"given":"Zhuoquan","family":"Yu","sequence":"first","affiliation":[{"name":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"}]},{"given":"Huidong","family":"Ji","sequence":"additional","affiliation":[{"name":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"}]},{"given":"Yue","family":"Cao","sequence":"additional","affiliation":[{"name":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"}]},{"given":"Junfu","family":"Wu","sequence":"additional","affiliation":[{"name":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"}]},{"given":"Xiaoze","family":"Yan","sequence":"additional","affiliation":[{"name":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"}]},{"given":"Lirong","family":"Zheng","sequence":"additional","affiliation":[{"name":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"}]},{"given":"Zhuo","family":"Zou","sequence":"additional","affiliation":[{"name":"Fudan University,State Key Laboratory of Integrated Chips and Systems,Shanghai,China"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Opt: Open pre-trained transformer language models","author":"Zhang","year":"2022","journal-title":"arXiv preprint arXiv:2205.01068"},{"key":"ref2","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint arXiv:2302.13971"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA47549.2020.00035"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/MWSCAS60917.2024.10658754"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/3453688.3461739"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3649329.3656221"},{"key":"ref7","article-title":"Omniquant: Omnidirectionally calibrated quantization for large language models","volume-title":"The Twelfth International Conference on Learning Representations","author":"Shao"},{"key":"ref8","first-page":"196","article-title":"Atom: Low-bit quantization for efficient and accurate 11m serving","volume-title":"Proceedings of Machine Learning and Systems","volume":"6","author":"Zhao"},{"key":"ref9","article-title":"Smoothquant: accurate and efficient post-training quantization for large language models","volume-title":"Proceedings of the 40th International Conference on Machine Learning, ICML\u201923","author":"Xiao"},{"key":"ref10","article-title":"Quarot: Outlier-free 4-bit inference in rotated llms","author":"Ashkboos","year":"2024","journal-title":"arXiv preprint arXiv:2404.00456"},{"key":"ref11","article-title":"Spinquant- 11 m quantization with learned rotations","author":"Liu","year":"2024","journal-title":"arXiv preprint arXiv:2405.16406"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO50266.2020.00071"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00060"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2024.3350661"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO56248.2022.00051"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1145\/3626202.3637562"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1145\/3656177"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2024.3487781"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3579371.3589038"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2021.3056337"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/FPL.2018.00063"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/FPL64840.2024.00044"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ASP-DAC58780.2024.10473976"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/FPL57034.2022.00035"},{"key":"ref25","article-title":"Massive activations in large language models","author":"Sun","year":"2024","journal-title":"arXiv preprint arXiv:2402.17762"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01549"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1145\/3640542"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3489517.3530505"},{"key":"ref29","article-title":"Pointer sentinel mixture models","author":"Merity","year":"2016","journal-title":"arXiv preprint arXiv:1609.07843"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i05.6239"},{"key":"ref31","article-title":"Think you have solved question answering? try arc, the ai2 reasoning challenge","author":"Clark","year":"2018","journal-title":"arXiv preprint arXiv:1803.05457"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1472"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1145\/3474381"},{"key":"ref34","article-title":"Deep learning with int8 optimization on xilinx devices white paper (wp.486)","author":"Fu","year":"2024"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.23919\/DATE.2017.7927113"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.23919\/fpl.2017.8056863"},{"key":"ref37","article-title":"Convolutional neural network with int4 optimization on xilinx devices (wp.521)","author":"Han","year":"2024"}],"event":{"name":"2025 62nd ACM\/IEEE Design Automation Conference (DAC)","location":"San Francisco, CA, USA","start":{"date-parts":[[2025,6,22]]},"end":{"date-parts":[[2025,6,25]]}},"container-title":["2025 62nd ACM\/IEEE Design Automation Conference (DAC)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11132383\/11132091\/11132816.pdf?arnumber=11132816","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,16]],"date-time":"2025-09-16T05:24:17Z","timestamp":1758000257000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11132816\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,22]]},"references-count":37,"URL":"https:\/\/doi.org\/10.1109\/dac63849.2025.11132816","relation":{},"subject":[],"published":{"date-parts":[[2025,6,22]]}}}