{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,26]],"date-time":"2026-02-26T15:23:56Z","timestamp":1772119436084,"version":"3.50.1"},"reference-count":50,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"4","license":[{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,4,1]],"date-time":"2025-04-01T00:00:00Z","timestamp":1743465600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100002367","name":"Strategic Priority Research Program of the Chinese Academy of Sciences","doi-asserted-by":"publisher","award":["XDB44000000"],"award-info":[{"award-number":["XDB44000000"]}],"id":[{"id":"10.13039\/501100002367","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. VLSI Syst."],"published-print":{"date-parts":[[2025,4]]},"DOI":"10.1109\/tvlsi.2025.3526363","type":"journal-article","created":{"date-parts":[[2025,1,15]],"date-time":"2025-01-15T20:32:40Z","timestamp":1736973160000},"page":"990-1003","source":"Crossref","is-referenced-by-count":3,"title":["SysCIM: A Heterogeneous Chip Architecture for High-Efficiency CNN Training at Edge"],"prefix":"10.1109","volume":"33","author":[{"ORCID":"https:\/\/orcid.org\/0009-0001-3730-4652","authenticated-orcid":false,"given":"Shuai","family":"Wang","sequence":"first","affiliation":[{"name":"School of Microelectronics, University of Science and Technology of China, Hefei, Anhui, China"}]},{"given":"Ziwei","family":"Li","sequence":"additional","affiliation":[{"name":"School of Microelectronics, University of Science and Technology of China, Hefei, Anhui, China"}]},{"given":"Yuang","family":"Ma","sequence":"additional","affiliation":[{"name":"School of Microelectronics, University of Science and Technology of China, Hefei, Anhui, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5487-6855","authenticated-orcid":false,"given":"Yi","family":"Kang","sequence":"additional","affiliation":[{"name":"School of Microelectronics, University of Science and Technology of China, Hefei, Anhui, China"}]}],"member":"263","reference":[{"key":"ref1","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. NIPS","author":"Brown"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.23919\/VLSICircuits52068.2021.9492476"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00240"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i4.16462"},{"key":"ref5","first-page":"5145","article-title":"Scalable methods for 8-bit training of neural networks","volume-title":"Proc. NIPS","volume":"31","author":"Banner"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/A-SSCC56115.2022.9980672"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/isscc42613.2021.9365766"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2021.3061508"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42614.2022.9731754"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i2.20039"},{"key":"ref11","first-page":"4107","article-title":"Binarized neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"29","author":"Hubara"},{"key":"ref12","first-page":"7474","article-title":"Rotated binary neural network","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Lin"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00232"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58568-6_9"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00777"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_32"},{"key":"ref17","first-page":"1742","article-title":"Flexpoint: An adaptive numerical format for efficient training of deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"K\u00f6ster"},{"key":"ref18","first-page":"1","article-title":"Mixed precision training","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Micikevicius"},{"key":"ref19","first-page":"4900","article-title":"Hybrid 8-bit floating point (HFP8) training and inference for deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Sun"},{"key":"ref20","first-page":"1","article-title":"A block minifloat representation for training deep neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Fox"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ISCA52012.2021.00061"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2019.2963616"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2022.3152653"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2020.2987714"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/isscc42614.2022.9731545"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.23919\/VLSICircuits52068.2021.9492444"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42614.2022.9731670"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/isscc42614.2022.9731621"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC42614.2022.9731715"},{"key":"ref30","first-page":"1","article-title":"A 40 nm 64 kb 26.56 TOPS\/W 2.37 Mb\/mm2RRAM binary\/compute-in-memory macro with 4.23\u00d7 improvement in density and >75% use of sensing dynamic range","volume-title":"IEEE Int. Solid-State Circuits Conf. (ISSCC) Dig. Tech. Papers","volume":"65","author":"Spetalnick"},{"key":"ref31","article-title":"DoReFa-Net: Training low bitwidth convolutional neural networks with low bitwidth gradients","author":"Zhou","year":"2016","journal-title":"arXiv:1606.06160"},{"key":"ref32","first-page":"1","article-title":"Training and inference with integers in deep neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Wu"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2019.12.027"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00204"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2023.126269"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2022.3140753"},{"key":"ref37","first-page":"246","article-title":"A 28 nm 64Kb 6T SRAM computing-in-memory macro with 8b MAC operation for AI edge chips","volume-title":"IEEE Int. Solid-State Circuits Conf. (ISSCC) Dig. Tech. Papers","author":"Si"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/TCSI.2023.3244338"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/JSSC.2021.3108344"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18072.2020.9218524"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC.2014.6757323"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ISCAS58744.2024.10558546"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1007\/0-387-21769-x_9"},{"key":"ref44","first-page":"84","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"25","author":"Krizhevsky"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/VLSIC.2018.8502276"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/HPCA53966.2022.00067"},{"key":"ref47","first-page":"144","article-title":"A 7 nm 4-core AI chip with 25.6TFLOPS hybrid FP8 training, 102.4TOPS INT4 inference and workload-aware throttling","volume-title":"IEEE Int. Solid-State Circuits Conf. (ISSCC) Dig. Tech. Papers","author":"Agrawal"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/ISSCC.2019.8662302"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/VLSICircuits18222.2020.9162917"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/EMC2-NIPS53020.2019.00010"}],"container-title":["IEEE Transactions on Very Large Scale Integration (VLSI) Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/92\/10937162\/10843320.pdf?arnumber=10843320","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,3,24]],"date-time":"2025-03-24T19:16:24Z","timestamp":1742843784000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10843320\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,4]]},"references-count":50,"journal-issue":{"issue":"4"},"URL":"https:\/\/doi.org\/10.1109\/tvlsi.2025.3526363","relation":{},"ISSN":["1063-8210","1557-9999"],"issn-type":[{"value":"1063-8210","type":"print"},{"value":"1557-9999","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,4]]}}}