{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T16:04:05Z","timestamp":1775837045849,"version":"3.50.1"},"reference-count":32,"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"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,6,22]]},"DOI":"10.1109\/dac63849.2025.11132777","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":["Mixed-Precision Quantization for Deep Vision Models with Integer Quadratic Programming"],"prefix":"10.1109","author":[{"given":"Zihao","family":"Deng","sequence":"first","affiliation":[{"name":"The University of Texas at Austin"}]},{"given":"Sayeh","family":"Sharify","sequence":"additional","affiliation":[{"name":"d-Matrix Corp"}]},{"given":"Xin","family":"Wang","sequence":"additional","affiliation":[{"name":"d-Matrix Corp"}]},{"given":"Michael","family":"Orshansky","sequence":"additional","affiliation":[{"name":"The University of Texas at Austin"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.23919\/DATE58400.2024.10546757"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01318"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW60793.2023.00144"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00530"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i10.29045"},{"key":"ref6","article-title":"BinaryConnect: Training deep neural networks with binary weights during propagations","volume":"28","author":"Courbariaux","year":"2015","journal-title":"Advances in neural information processing systems"},{"key":"ref7","article-title":"Cvxpy: A python-embedded modeling language for convex optimization","volume":"17","author":"Diamond","year":"2016","journal-title":"Journal of machine learning research: JMLR"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01566"},{"key":"ref9","article-title":"HAWQ-V2: Hessian aware trace-weighted quantization of neural networks","author":"Dong","year":"2020","journal-title":"ArXiv, abs\/1911.03852"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00038"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2010.11929"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58517-4_32"},{"key":"ref13","article-title":"Gurobi Optimization, LLC. Gurobi Optimizer Reference Manual","year":"2023"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/iccv43118.2019"},{"key":"ref16","article-title":"Integer-only bert quantization","author":"Kim","year":"2021","journal-title":"ArXiv, abs\/2101.01321"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.23919\/DATE54114.2022.9774740"},{"key":"ref18","article-title":"BRECQ: Pushing the limit of post-training quantization by block reconstruction","author":"Li","year":"2021","journal-title":"ArXiv, abs\/2102.05426"},{"key":"ref19","article-title":"Mqbench: Towards reproducible and deployable model quantization benchmark","volume-title":"Proceedings of the Neural Information Processing Systems Track on Datasets and Benchmarks","author":"Li"},{"key":"ref20","article-title":"AutoQ: Automated kernel-wise neural network quantization","author":"Lou","year":"2020","journal-title":"ArXiv, abs\/1902.05690"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i7.26084"},{"key":"ref22","article-title":"Up or down? adaptive rounding for post-training quantization","author":"Nagel","year":"2020","journal-title":"ArXiv, abs\/2004.10568"},{"key":"ref23","article-title":"PyTorch: An imperative style, high-performance deep learning library","volume":"32","author":"Paszke","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref24","first-page":"87488763","article-title":"Learning transferable visual models from natural language supervision","volume-title":"International conference on machine learning","author":"Radford"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01044"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20083-0_16"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00881"},{"key":"ref29","article-title":"QDrop: Randomly dropping quantization for extremely low-bit post-training quantization","author":"Wei","year":"2022","journal-title":"ArXiv, abs\/2203.05740"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-demos.6"},{"key":"ref31","article-title":"Mixed precision quantization of convnets via differentiable neural architecture search","author":"Wu","year":"2018","journal-title":"ArXiv, abs\/1812.00090"},{"key":"ref32","first-page":"11875","article-title":"HAWQ-V3: Dyadic neural network quantization","volume-title":"International Conference on Machine Learning","author":"Yao"}],"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\/11132777.pdf?arnumber=11132777","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,16]],"date-time":"2025-09-16T05:24:22Z","timestamp":1758000262000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11132777\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,22]]},"references-count":32,"URL":"https:\/\/doi.org\/10.1109\/dac63849.2025.11132777","relation":{},"subject":[],"published":{"date-parts":[[2025,6,22]]}}}