{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,7,23]],"date-time":"2026-07-23T15:34:30Z","timestamp":1784820870770,"version":"3.55.0"},"reference-count":64,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"},{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"NSF","award":["DMS-2208126"],"award-info":[{"award-number":["DMS-2208126"]}]},{"name":"NSF","award":["IIS-2110546"],"award-info":[{"award-number":["IIS-2110546"]}]},{"name":"NSF","award":["DMS-2219904"],"award-info":[{"award-number":["DMS-2219904"]}]},{"name":"NSF","award":["DMS-2151225"],"award-info":[{"award-number":["DMS-2151225"]}]},{"name":"NSF","award":["DMS2219904"],"award-info":[{"award-number":["DMS2219904"]}]},{"name":"NSF","award":["CCSS-2348046"],"award-info":[{"award-number":["CCSS-2348046"]}]},{"name":"Start-Up"},{"DOI":"10.13039\/100009790","name":"University at Albany","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100009790","id-type":"DOI","asserted-by":"publisher"}]},{"name":"SUNY-IBM AI Collaborative Research"},{"name":"Qualcomm Faculty Gift Award"},{"name":"NSF","award":["DMS-2208126"],"award-info":[{"award-number":["DMS-2208126"]}]},{"name":"NSF","award":["IIS-2110546"],"award-info":[{"award-number":["IIS-2110546"]}]},{"name":"SUNY-IBM AI Research Alliance Grant"},{"name":"NSF","award":["DMS-2151225"],"award-info":[{"award-number":["DMS-2151225"]}]},{"name":"NSF","award":["DMS-2219904"],"award-info":[{"award-number":["DMS-2219904"]}]},{"name":"Qualcomm Faculty Gift Award"},{"name":"NSF","award":["CCSS-2348046"],"award-info":[{"award-number":["CCSS-2348046"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Access"],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/access.2025.3576737","type":"journal-article","created":{"date-parts":[[2025,6,6]],"date-time":"2025-06-06T17:43:06Z","timestamp":1749231786000},"page":"99879-99891","source":"Crossref","is-referenced-by-count":4,"title":["COMQ: A Backpropagation-Free Algorithm for Post-Training Quantization"],"prefix":"10.1109","volume":"13","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6292-1348","authenticated-orcid":false,"given":"Aozhong","family":"Zhang","sequence":"first","affiliation":[{"name":"Department of Mathematics and Statistics, State University of New York at Albany, Albany, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zi","family":"Yang","sequence":"additional","affiliation":[{"name":"Department of Mathematics and Statistics, State University of New York at Albany, Albany, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7664-0061","authenticated-orcid":false,"given":"Naigang","family":"Wang","sequence":"additional","affiliation":[{"name":"IBM T. J. Watson Research Center, Yorktown Heights, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yingyong","family":"Qi","sequence":"additional","affiliation":[{"name":"Department of Mathematics, University of California at Irvine, Irvine, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6438-8476","authenticated-orcid":false,"given":"Jack","family":"Xin","sequence":"additional","affiliation":[{"name":"Department of Mathematics, University of California at Irvine, Irvine, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2067-2763","authenticated-orcid":false,"given":"Xin","family":"Li","sequence":"additional","affiliation":[{"name":"Department of Computer Science, State University of New York at Albany, Albany, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6098-8493","authenticated-orcid":false,"given":"Penghang","family":"Yin","sequence":"additional","affiliation":[{"name":"Department of Mathematics and Statistics, State University of New York at Albany, Albany, NY, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2016.2577031"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.13140\/RG.2.2.18893.74727"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-021-03819-2"},{"key":"ref5","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv:1810.04805"},{"key":"ref6","article-title":"Improving language understanding by generative pre-training","volume-title":"OpenAI","author":"Radford","year":"2018"},{"key":"ref7","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ho"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00206"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.155"},{"key":"ref10","article-title":"Pruning filters for efficient ConvNets","author":"Li","year":"2016","journal-title":"arXiv:1608.08710"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_32"},{"key":"ref12","article-title":"Mixed precision DNNs: All you need is a good parametrization","author":"Uhlich","year":"2019","journal-title":"arXiv:1905.11452"},{"key":"ref13","article-title":"DoReFa-net: Training low bitwidth convolutional neural networks with low bitwidth gradients","author":"Zhou","year":"2016","journal-title":"arXiv:1606.06160"},{"issue":"1","key":"ref14","first-page":"6869","article-title":"Quantized neural networks: Training neural networks with low precision weights and activations","volume":"18","author":"Hubara","year":"2016","journal-title":"J. Mach. Learn. Res."},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr.2017.422"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1007\/s40687-018-0177-6"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1137\/18M1166134"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref19","article-title":"Understanding straight-through estimator in training activation quantized neural nets","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Yin"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3297890"},{"key":"ref21","article-title":"Quantization and training of low bit-width convolutional neural networks for object detection","author":"Yin","year":"2016","journal-title":"arXiv:1612.06052"},{"key":"ref22","article-title":"BRECQ: Pushing the limit of post-training quantization by block reconstruction","author":"Li","year":"2021","journal-title":"arXiv:2102.05426"},{"key":"ref23","first-page":"7197","article-title":"Up or down? Adaptive rounding for post-training quantization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Nagel"},{"key":"ref24","first-page":"4475","article-title":"Optimal brain compression: A framework for accurate post-training quantization and pruning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Frantar"},{"key":"ref25","first-page":"16196","article-title":"Bit-shrinking: Limiting instantaneous sharpness for improving post-training quantization","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Chen"},{"key":"ref26","article-title":"Improving post training neural quantization: Layer-wise calibration and integer programming","author":"Hubara","year":"2020","journal-title":"arXiv:2006.10518"},{"key":"ref27","article-title":"QDrop: Randomly dropping quantization for extremely low-bit post-training quantization","author":"Wei","year":"2022","journal-title":"arXiv:2203.05740"},{"key":"ref28","first-page":"14140","article-title":"Deep compression of pre-trained transformer models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Wang"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2010.11929"},{"key":"ref30","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Touvron"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref32","first-page":"28092","article-title":"Post-training quantization for vision transformer","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liu"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1137\/22M1511709"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3159369"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1137\/100802001"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/s10107-015-0892-3"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1145\/1390156.1390208"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01580"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00141"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01318"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58610-2_1"},{"key":"ref42","first-page":"4466","article-title":"Accurate post training quantization with small calibration sets","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Hubara"},{"key":"ref43","article-title":"GPTQ: Accurate post-training quantization for generative pre-trained transformers","author":"Frantar","year":"2022","journal-title":"arXiv:2210.17323"},{"key":"ref44","article-title":"QuantEase: Optimization-based quantization for language models","author":"Behdin","year":"2023","journal-title":"arXiv:2309.01885"},{"key":"ref45","article-title":"PTQ4 ViT: Post-training quantization for vision transformers with twin uniform quantization","author":"Yuan","year":"2021","journal-title":"arXiv:2111.12293"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3547826"},{"key":"ref47","article-title":"FQ-ViT: Post-training quantization for fully quantized vision transformer","author":"Lin","year":"2021","journal-title":"arXiv:2111.13824"},{"key":"ref48","article-title":"PTQD: Accurate post-training quantization for diffusion models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"He"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00196"},{"key":"ref50","first-page":"38087","article-title":"SmoothQuant: Accurate and efficient post-training quantization for large language models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Xiao"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref52","article-title":"PACT: Parameterized clipping activation for quantized neural networks","author":"Choi","year":"2018","journal-title":"arXiv:1805.06085"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref54","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017","journal-title":"arXiv:1704.04861"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref56","article-title":"OPT: Open pre-trained transformer language models","author":"Zhang","year":"2022","journal-title":"arXiv:2205.01068"},{"key":"ref57","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv:2307.09288"},{"key":"ref58","article-title":"Pointer sentinel mixture models","author":"Merity","year":"2016","journal-title":"arXiv:1609.07843"},{"key":"ref59","first-page":"9847","article-title":"Towards accurate post-training network quantization via bit-split and stitching","volume-title":"Proc. Int. Conf. Mach. Learn.","volume":"1","author":"Wang"},{"key":"ref60","article-title":"FlexRound: Learnable rounding based on element-wise division for post-training quantization","author":"Lee","year":"2023","journal-title":"arXiv:2306.00317"},{"key":"ref61","first-page":"1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2019","journal-title":"J. Mach. Learn. Res."},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02340"},{"key":"ref63","article-title":"Stable and low-precision training for large-scale vision-language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wortsman"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW60793.2023.00144"}],"container-title":["IEEE Access"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/6287639\/10820123\/11026044-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6287639\/10820123\/11026044.pdf?arnumber=11026044","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,13]],"date-time":"2025-06-13T17:53:06Z","timestamp":1749837186000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11026044\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":64,"URL":"https:\/\/doi.org\/10.1109\/access.2025.3576737","relation":{},"ISSN":["2169-3536"],"issn-type":[{"value":"2169-3536","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025]]}}}