{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T15:50:11Z","timestamp":1781884211272,"version":"3.54.5"},"reference-count":80,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"7","license":[{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,7,1]],"date-time":"2025-07-01T00:00:00Z","timestamp":1751328000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Beijing Municipal Science and Technology Project","award":["Z231100010323002"],"award-info":[{"award-number":["Z231100010323002"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2025,7]]},"DOI":"10.1109\/tpami.2025.3554523","type":"journal-article","created":{"date-parts":[[2025,3,26]],"date-time":"2025-03-26T07:28:13Z","timestamp":1742974093000},"page":"5556-5570","source":"Crossref","is-referenced-by-count":5,"title":["Pushing the Limit of Post-Training Quantization"],"prefix":"10.1109","volume":"47","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6024-7086","authenticated-orcid":false,"given":"Ruihao","family":"Gong","sequence":"first","affiliation":[{"name":"State Key Laboratory of Complex & Critical Software Environment, Beihang University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8425-4195","authenticated-orcid":false,"given":"Xianglong","family":"Liu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Complex & Critical Software Environment, Beihang University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6444-7253","authenticated-orcid":false,"given":"Yuhang","family":"Li","sequence":"additional","affiliation":[{"name":"Yale University, New Haven, CT, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yunqiang","family":"Fan","sequence":"additional","affiliation":[{"name":"ShanghaiTech University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiuying","family":"Wei","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Complex & Critical Software Environment, Beihang University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1956-3367","authenticated-orcid":false,"given":"Jinyang","family":"Guo","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Complex & Critical Software Environment, Beihang University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01044"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.236"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW54120.2021.00107"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1810.04805"},{"key":"ref6","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"key":"ref7","first-page":"1","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":"ref8","first-page":"1","article-title":"Neural architecture search with reinforcement learning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zoph"},{"issue":"1","key":"ref9","first-page":"6869","article-title":"Quantized neural networks: Training neural networks with low precision weights and activations","volume":"18","author":"Hubara","year":"2017","journal-title":"J. Mach. Learn. Res."},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00286"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00703"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.102"},{"key":"ref13","first-page":"1","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Han"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00158"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i07.6720"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2020.2996231"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i11.29108"},{"key":"ref18","first-page":"1","article-title":"Learned step size quantization","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Esser"},{"key":"ref19","first-page":"348","article-title":"Accurate and efficient 2-bit quantized neural networks","volume-title":"Proc. Mach. Learn. Syst.","author":"Choi"},{"key":"ref20","first-page":"1","article-title":"Additive powers-of-two quantization: An efficient non-uniform discretization for neural networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Li"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00495"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00141"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01540"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00437"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01318"},{"key":"ref26","first-page":"9847","article-title":"Towards accurate post-training network quantization via bit-split and stitching","volume-title":"Proc. 37nd Int. Conf. Mach. Learn.","author":"Wang"},{"key":"ref27","first-page":"4466","article-title":"Accurate post training quantization with small calibration sets","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Hubara"},{"key":"ref28","first-page":"7950","article-title":"Post training 4-bit quantization of convolutional networks for rapid-deployment","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Banner"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-021-06053-z"},{"key":"ref30","first-page":"7197","article-title":"Up or down? Adaptive rounding for post-training quantization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Nagel"},{"key":"ref31","first-page":"1","article-title":"BRECQ: Pushing the limit of post-training quantization by block reconstruction","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Li"},{"key":"ref32","first-page":"1","article-title":"QDrop: Randomly dropping quantization for extremely low-bit post-training quantization","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Wei"},{"key":"ref33","article-title":"Estimating or propagating gradients through stochastic neurons for conditional computation","author":"Bengio","year":"2013"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01237-3_23"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00499"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00881"},{"key":"ref37","first-page":"18518","article-title":"HAWQ-V2: Hessian aware trace-weighted quantization of neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Dong"},{"key":"ref38","first-page":"164","article-title":"Second order derivatives for network pruning: Optimal brain surgeon","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Hassibi"},{"key":"ref39","first-page":"4860","article-title":"Learning to prune deep neural networks via layer-wise optimal brain surgeon","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Dong"},{"key":"ref40","first-page":"557","article-title":"Practical gauss-newton optimisation for deep learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Botev"},{"key":"ref41","first-page":"735","article-title":"Deep learning via hessian-free optimization","volume-title":"Proc. 27th Int. Conf. Int. Conf. Mach. Learn.","author":"Martens"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1162\/089976698300017746"},{"key":"ref43","first-page":"2408","article-title":"Optimizing neural networks with kronecker-factored approximate curvature","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Martens"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.1.1"},{"key":"ref45","first-page":"1","article-title":"Fantastic generalization measures and where to find them","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Jiang"},{"key":"ref46","first-page":"1","article-title":"On large-batch training for deep learning: Generalization gap and sharp minima","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Keskar"},{"key":"ref47","first-page":"2958","article-title":"Adversarial weight perturbation helps robust generalization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wu"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00806"},{"key":"ref49","first-page":"876","article-title":"Averaging weights leads to wider optima and better generalization","volume-title":"Proc. 34th Conf. Uncertainty Artif. Intell.","author":"Izmailov"},{"key":"ref50","first-page":"1","article-title":"Sharpness-aware minimization for efficiently improving generalization","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Foret"},{"key":"ref51","first-page":"7015","article-title":"SWALP: Stochastic weight averaging in low precision training","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Yang"},{"key":"ref52","first-page":"1","article-title":"Comparing fisher information regularization with distillation for DNN quantization","volume-title":"Proc. Workshop Deep Learn. Through Inf. Geometry","author":"Kadambi"},{"key":"ref53","first-page":"1","article-title":"Training with quantization noise for extreme model compression","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Stock"},{"key":"ref54","first-page":"5949","article-title":"Exploring generalization in deep learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Neyshabur"},{"key":"ref55","first-page":"1","article-title":"Distilling the knowledge in a neural network","volume-title":"Proc. 2014 Deep Learn. Workshop","author":"Hinton"},{"key":"ref56","first-page":"1019","article-title":"Sharp minima can generalize for deep nets","volume-title":"Proc. 34th Int. Conf. Mach. Learn.","author":"Dinh"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00232"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2019.00363"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/W18-5446"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/D16-1264"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00293"},{"key":"ref65","first-page":"1","article-title":"An image is worth 16 \u00d7 16 words: Transformers for image recognition at scale","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Dosovitskiy"},{"key":"ref66","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Touvron"},{"key":"ref67","first-page":"28092","article-title":"Post-training quantization for vision transformer","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liu"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19775-8_12"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2016.2577031"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/EMC2-NIPS53020.2019.00016"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.334"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW53098.2021.00339"},{"key":"ref74","first-page":"1","article-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009"},{"key":"ref75","first-page":"1","article-title":"Adam: A method for stochastic optimization","volume-title":"Proc. 3rd Int. Conf. Learn. Representations","author":"Kingma"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58517-4_32"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA.2018.00069"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1145\/3404397.3404407"},{"key":"ref79","first-page":"1","article-title":"MQBench: Towards reproducible and deployable model quantization benchmark","volume-title":"Proc. 35th Conf. Neural Inf. Process. Syst. Datasets Benchmarks Track","author":"Li"},{"key":"ref80","first-page":"3492","article-title":"Double-win quant: Aggressively winning robustness of quantized deep neural networks via random precision training and inference","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Fu"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11026037\/10938638.pdf?arnumber=10938638","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,6]],"date-time":"2025-06-06T04:19:37Z","timestamp":1749183577000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10938638\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7]]},"references-count":80,"journal-issue":{"issue":"7"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2025.3554523","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,7]]}}}