{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T09:37:30Z","timestamp":1777887450889,"version":"3.51.4"},"reference-count":43,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"5","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100005089","name":"Beijing Natural Science Foundation","doi-asserted-by":"publisher","award":["L243031"],"award-info":[{"award-number":["L243031"]}],"id":[{"id":"10.13039\/501100005089","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Key Research and Development Program of China","award":["2023YFB4503704"],"award-info":[{"award-number":["2023YFB4503704"]}]},{"name":"National Key Research and Development Program of China","award":["2024YFB4505601"],"award-info":[{"award-number":["2024YFB4505601"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62572036"],"award-info":[{"award-number":["62572036"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Circuits Syst. I"],"published-print":{"date-parts":[[2026,5]]},"DOI":"10.1109\/tcsi.2025.3638448","type":"journal-article","created":{"date-parts":[[2025,12,3]],"date-time":"2025-12-03T18:45:38Z","timestamp":1764787538000},"page":"3350-3362","source":"Crossref","is-referenced-by-count":0,"title":["TinyFormer: Efficient Sparse Transformer Design and Deployment on Tiny Devices"],"prefix":"10.1109","volume":"73","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-8424-7040","authenticated-orcid":false,"given":"Jianlei","family":"Yang","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiacheng","family":"Liao","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Fanding","family":"Lei","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Meichen","family":"Liu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Lingkun","family":"Long","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0003-7397-6311","authenticated-orcid":false,"given":"Junyi","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9238-451X","authenticated-orcid":false,"given":"Han","family":"Wan","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6406-4810","authenticated-orcid":false,"given":"Bei","family":"Yu","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Engineering, The Chinese University of Hong Kong, Hong Kong, SAR, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8088-0404","authenticated-orcid":false,"given":"Weisheng","family":"Zhao","sequence":"additional","affiliation":[{"name":"Fert Beijing Research Institute, School of Integrated Circuit Science and Engineering, Beihang University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18074.2021.9586232"},{"key":"ref2","article-title":"Hello edge: Keyword spotting on microcontrollers","author":"Zhang","year":"2017","journal-title":"arXiv:1711.07128"},{"key":"ref3","article-title":"TinyML: Tools, applications, challenges, and future research directions","author":"Kallimani","year":"2023","journal-title":"arXiv:2303.13569"},{"key":"ref4","article-title":"Benchmarking TinyML systems: Challenges and direction","author":"Banbury","year":"2020","journal-title":"arXiv:2003.04821"},{"key":"ref5","first-page":"517","article-title":"MicroNets: Neural network architectures for deploying TinyML applications on commodity microcontrollers","volume-title":"Proc. Int. Conf. Mach. Learn. Syst.","author":"Banbury"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-3015"},{"key":"ref9","first-page":"1","article-title":"An image is worth 16$\\times$\n16 words: Transformers for image recognition at scale","volume-title":"Proc. 9th Int. Conf. Learn. Represent. (ICLR)","author":"Dosovitskiy"},{"key":"ref10","first-page":"1","article-title":"MobileViT: Light-weight, general-purpose, and mobile-friendly vision transformer","volume-title":"Proc. 10th Int. Conf. Learn. Represent. (ICLR)","author":"Mehta"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19809-0_35"},{"key":"ref12","article-title":"Escaping the big data paradigm with compact transformers","author":"Hassani","year":"2021","journal-title":"arXiv:2104.05704"},{"key":"ref13","article-title":"CMSIS-NN: Efficient neural network kernels for arm cortex-M CPUs","author":"Lai","year":"2018","journal-title":"arXiv:1801.06601"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.23919\/DATE58400.2024.10546541"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-66770-2_22"},{"key":"ref16","first-page":"11711","article-title":"MCUNet: Tiny deep learning on IoT devices","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NeurIPS)","author":"Lin"},{"key":"ref17","article-title":"MCUNetV2: Memory-efficient patch-based inference for tiny deep learning","author":"Lin","year":"2021","journal-title":"arXiv:2110.15352"},{"key":"ref18","first-page":"1","article-title":"On-device training under 256KB memory","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NeurIPS)","author":"Lin"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/COINS51742.2021.9524173"},{"key":"ref20","first-page":"1","article-title":"Mcuformer: Deploying vision tranformers on microcontrollers with limited memory","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NeurIPS)","author":"Yinan"},{"key":"ref21","first-page":"1","article-title":"Neural architecture search with reinforcement learning","volume-title":"Proc. Int. Conf. Learn. Represent. (ICLR)","author":"Zoph"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/DAC56929.2023.10247875"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01099"},{"key":"ref24","first-page":"1","article-title":"ProxylessNAS: Direct neural architecture search on target task and hardware","volume-title":"Proc. 7th Int. Conf. Learn. Represent. (ICLR)","author":"Cai"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TCAD.2020.2986127"},{"key":"ref26","first-page":"4978","article-title":"SpArSe: Sparse architecture search for CNNs on resource-constrained microcontrollers","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NeurIPS)","author":"Fedorov"},{"key":"ref27","first-page":"265","article-title":"TensorFlow: A system for large-scale machine learning","volume-title":"Proc. 12th USENIX Symp. Operating Syst. Design Implement. (OSDI)","author":"Abadi"},{"key":"ref28","first-page":"578","article-title":"TVM: An automated end-to-end optimizing compiler for deep learning","volume-title":"Proc. 13th USENIX Symp. Operating Syst. Design Implement. (OSDI)","author":"Chen"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/TCSII.2020.2983648"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58517-4_32"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref32","article-title":"Deep learning using rectified linear units (ReLU)","author":"Fred Agarap","year":"2018","journal-title":"arXiv:1803.08375"},{"key":"ref33","first-page":"1","article-title":"To prune, or not to prune: Exploring the efficacy of pruning for model compression","volume-title":"Proc. 6th Int. Conf. Learn. Represent. (ICLR)","author":"Zhu"},{"key":"ref34","article-title":"Quantizing deep convolutional networks for efficient inference: A whitepaper","author":"Krishnamoorthi","year":"2018","journal-title":"arXiv:1806.08342"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/PROC.1967.5493"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2021.3087946"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/DAC18072.2020.9218710"},{"key":"ref38","first-page":"1","article-title":"Separable self-attention for mobile vision transformers","volume":"2023","author":"Mehta","year":"2023","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref39","article-title":"Learning multiple layers of features from tiny images","author":"Krizhevsky","year":"2009"},{"key":"ref40","article-title":"A downsampled variant of ImageNet as an alternative to the CIFAR datasets","author":"Chrabaszcz","year":"2017","journal-title":"arXiv:1707.08819"},{"key":"ref41","first-page":"1","article-title":"Decoupled weight decay regularization","volume-title":"Proc. 7th Int. Conf. Learn. Represent. (ICLR)","author":"Loshchilov"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"}],"container-title":["IEEE Transactions on Circuits and Systems I: Regular Papers"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/8919\/11500576\/11275632.pdf?arnumber=11275632","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T04:13:09Z","timestamp":1777608789000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11275632\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":43,"journal-issue":{"issue":"5"},"URL":"https:\/\/doi.org\/10.1109\/tcsi.2025.3638448","relation":{},"ISSN":["1549-8328","1558-0806"],"issn-type":[{"value":"1549-8328","type":"print"},{"value":"1558-0806","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5]]}}}