{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T21:15:41Z","timestamp":1765228541072,"version":"3.46.0"},"reference-count":79,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"12","license":[{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Flemish Government under the Onderzoeksprogramma Artifici\u00eble Intelligentie Vlaanderen Programme"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Circuits Syst. Video Technol."],"published-print":{"date-parts":[[2025,12]]},"DOI":"10.1109\/tcsvt.2025.3588737","type":"journal-article","created":{"date-parts":[[2025,7,14]],"date-time":"2025-07-14T17:44:27Z","timestamp":1752515067000},"page":"11950-11964","source":"Crossref","is-referenced-by-count":0,"title":["Improving Post-Training Quantization via Probabilistic Programming"],"prefix":"10.1109","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-2393-4362","authenticated-orcid":false,"given":"Kui","family":"Liu","sequence":"first","affiliation":[{"name":"Image Processing and Interpretation Research Group, Interuniversity Microelectronics Centre (IMEC), Ghent University, Ghent, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1666-5483","authenticated-orcid":false,"given":"Bart","family":"Goossens","sequence":"additional","affiliation":[{"name":"Image Processing and Interpretation Research Group, Interuniversity Microelectronics Centre (IMEC), Ghent University, Ghent, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2969-3133","authenticated-orcid":false,"given":"Tom","family":"De Schepper","sequence":"additional","affiliation":[{"name":"AI & Data Department, Interuniversity Microelectronics Centre (IMEC), Leuven, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4456-4353","authenticated-orcid":false,"given":"Wilfried","family":"Philips","sequence":"additional","affiliation":[{"name":"Image Processing and Interpretation Research Group, Interuniversity Microelectronics Centre (IMEC), Ghent University, Ghent, Belgium"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","author":"Han","year":"2015","journal-title":"arXiv:1510.00149"},{"key":"ref2","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017","journal-title":"arXiv:1704.04861"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1503.02531"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00286"},{"key":"ref5","article-title":"MeliusNet: Can binary neural networks achieve MobileNet-level accuracy?","author":"Bethge","year":"2020","journal-title":"arXiv:2001.05936"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2022.3216776"},{"key":"ref7","article-title":"PACT: Parameterized clipping activation for quantized neural networks","author":"Choi","year":"2018","journal-title":"arXiv:1805.06085"},{"key":"ref8","article-title":"Learned step size quantization","author":"Esser","year":"2019","journal-title":"arXiv:1902.08153"},{"key":"ref9","article-title":"A survey of quantization methods for efficient neural network inference","author":"Gholami","year":"2021","journal-title":"arXiv:2103.13630"},{"key":"ref10","first-page":"7948","article-title":"Post training 4-bit quantization of convolutional networks for rapid-deployment","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Banner"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2019.00363"},{"key":"ref12","first-page":"7197","article-title":"Up or down? Adaptive rounding for post-training quantization","volume-title":"Proc. Int. Conf. Machine. Learn. (ICML)","author":"Nagel"},{"key":"ref13","article-title":"BRECQ: Pushing the limit of post-training quantization by block reconstruction","author":"Li","year":"2021","journal-title":"arXiv:2102.05426"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-023-43715-5"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1038\/nature14541"},{"key":"ref16","article-title":"Deep probabilistic programming languages: A qualitative study","author":"Baudart","year":"2018","journal-title":"arXiv:1804.06458"},{"key":"ref17","first-page":"3290","article-title":"Bayesian compression for deep learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Louizos"},{"article-title":"Variational network quantization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Achterhold","key":"ref18"},{"key":"ref19","first-page":"5741","article-title":"Bayesian bits: Unifying quantization and pruning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Baalen"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.7717\/peerj-cs.55"},{"issue":"1","key":"ref21","first-page":"973","article-title":"Pyro: Deep universal probabilistic programming","volume":"20","author":"Bingham","year":"2019","journal-title":"J. Mach. Learn. Res."},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01237-3_23"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2024.3368959"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_32"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2022.3216389"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00141"},{"key":"ref27","first-page":"7543","article-title":"Improving neural network quantization without retraining using outlier channel splitting","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Zhao"},{"key":"ref28","first-page":"17737","article-title":"Post-training sparsity-aware quantization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Shomron"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3159369"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3311923"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3323015"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-021-06053-z"},{"key":"ref33","first-page":"4466","article-title":"Accurate post training quantization with small calibration sets","volume-title":"Proc. 38th Int. Conf. Mach. Learn.","author":"Hubara"},{"key":"ref34","article-title":"QDrop: Randomly dropping quantization for extremely low-bit post-training quantization","author":"Wei","year":"2022","journal-title":"arXiv:2203.05740"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02340"},{"key":"ref36","article-title":"SQuant: On-the-fly data-free quantization via diagonal Hessian approximation","author":"Guo","year":"2022","journal-title":"arXiv:2202.07471"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01318"},{"key":"ref38","first-page":"14835","article-title":"Qimera: Data-free quantization with synthetic boundary supporting samples","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Choi"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3279281"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3399311"},{"key":"ref41","article-title":"A white paper on neural network quantization","author":"Nagel","year":"2021","journal-title":"arXiv:2106.08295"},{"key":"ref42","article-title":"Quantizing deep convolutional networks for efficient inference: A whitepaper","author":"Krishnamoorthi","year":"2018","journal-title":"arXiv:1806.08342"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/19.492748"},{"key":"ref44","first-page":"12367","article-title":"Universally quantized neural compression","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Agustsson"},{"key":"ref45","article-title":"End-to-end optimized image compression","author":"Ball\u00e9","year":"2016","journal-title":"arXiv:1611.01704"},{"key":"ref46","first-page":"2617","article-title":"Variational autoencoder for low bit-rate image compression","volume-title":"Proc. IEEE Conf. Comput. Vis. Pattern Recognit. Workshops","author":"Zhou"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/18.720554"},{"key":"ref48","first-page":"2575","article-title":"Variational dropout and the local reparameterization trick","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"28","author":"Kingma"},{"key":"ref49","article-title":"TensorFlow distributions","author":"Dillon","year":"2017","journal-title":"arXiv:1711.10604"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1312.6114"},{"key":"ref51","first-page":"1278","article-title":"Stochastic backpropagation and approximate inference in deep generative models","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Rezende"},{"key":"ref52","first-page":"441","article-title":"Implicit reparameterization gradients","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Figurnov"},{"key":"ref53","article-title":"Joint distributions for TensorFlow probability","author":"Piponi","year":"2020","journal-title":"arXiv:2001.11819"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1017\/CBO9780511804441"},{"key":"ref55","article-title":"Drafting event schemas using language models","author":"Gunjal","year":"2023","journal-title":"arXiv:2305.14847"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/DICTA60407.2023.00025"},{"key":"ref57","first-page":"1070","article-title":"Early stopping as nonparametric variational inference","volume-title":"Proc. Artif. Intell. Statist.","author":"Duvenaud"},{"key":"ref58","first-page":"1","article-title":"Automatic termination for hyperparameter optimization","volume-title":"Proc. Int. Conf. Automated Mach. Learn.","author":"Makarova"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"article-title":"Learning multiple layers of features from tiny images","year":"2009","author":"Krizhevsky","key":"ref60"},{"key":"ref61","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2014","journal-title":"arXiv:1412.6980"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref65","article-title":"An image is worth 16\u00d716 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020","journal-title":"arXiv:2010.11929"},{"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","article-title":"GPTQ: Accurate post-training quantization for generative pre-trained transformers","author":"Frantar","year":"2022","journal-title":"arXiv:2210.17323"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01507"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01201"},{"key":"ref70","article-title":"CBQ: Cross-block quantization for large language models","author":"Ding","year":"2023","journal-title":"arXiv:2312.07950"},{"key":"ref71","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":"ref72","first-page":"17402","article-title":"Outlier suppression: Pushing the limit of low-bit transformer language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wei"},{"key":"ref73","article-title":"Outlier Suppression+: Accurate quantization of large language models by equivalent and optimal shifting and scaling","author":"Wei","year":"2023","journal-title":"arXiv:2304.09145"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19775-8_12"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3547826"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01580"},{"key":"ref77","article-title":"LRP-QViT: Mixed-precision vision transformer quantization via layer-wise relevance propagation","author":"Ranjan","year":"2024","journal-title":"arXiv:2401.11243"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2025.3528042"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00874"}],"container-title":["IEEE Transactions on Circuits and Systems for Video Technology"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/76\/11278843\/11080033.pdf?arnumber=11080033","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,8]],"date-time":"2025-12-08T18:42:41Z","timestamp":1765219361000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11080033\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12]]},"references-count":79,"journal-issue":{"issue":"12"},"URL":"https:\/\/doi.org\/10.1109\/tcsvt.2025.3588737","relation":{},"ISSN":["1051-8215","1558-2205"],"issn-type":[{"type":"print","value":"1051-8215"},{"type":"electronic","value":"1558-2205"}],"subject":[],"published":{"date-parts":[[2025,12]]}}}