{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,8,28]],"date-time":"2025-08-28T05:10:16Z","timestamp":1756357816278,"version":"3.44.0"},"reference-count":35,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,7,6]],"date-time":"2025-07-06T00:00:00Z","timestamp":1751760000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,7,6]],"date-time":"2025-07-06T00:00:00Z","timestamp":1751760000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,7,6]]},"DOI":"10.1109\/isvlsi65124.2025.11130246","type":"proceedings-article","created":{"date-parts":[[2025,8,27]],"date-time":"2025-08-27T18:20:15Z","timestamp":1756318815000},"page":"1-6","source":"Crossref","is-referenced-by-count":0,"title":["Dynamic Early-Exit Convolutional Neural Networks for Edge Vision: The Benefits, The Challenges, and the Road Ahead"],"prefix":"10.1109","author":[{"given":"Michalis","family":"Piponidis","sequence":"first","affiliation":[{"name":"University of Cyprus,Department of Electrical and Computer Engineering and KIOS Research and Innovation Center of Excellence,Nicosia,Cyprus"}]},{"given":"Georgios","family":"Konstantinidis","sequence":"additional","affiliation":[{"name":"University of Cyprus,Department of Electrical and Computer Engineering and KIOS Research and Innovation Center of Excellence,Nicosia,Cyprus"}]},{"given":"Maria K.","family":"Michael","sequence":"additional","affiliation":[{"name":"University of Cyprus,Department of Electrical and Computer Engineering and KIOS Research and Innovation Center of Excellence,Nicosia,Cyprus"}]},{"given":"Theocharis","family":"Theocharides","sequence":"additional","affiliation":[{"name":"University of Cyprus,Department of Electrical and Computer Engineering and KIOS Research and Innovation Center of Excellence,Nicosia,Cyprus"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3065386"},{"key":"ref2","first-page":"130313","article-title":"Dynamic inference in resource-efficient deep learning: A survey","volume":"9","author":"Qendro","year":"2021","journal-title":"IEEE Access"},{"article-title":"Slimmable neural networks","volume-title":"Proc. Int. Conf. Learn. Representations (ICLR)","author":"Yu","key":"ref3"},{"issue":"1","key":"ref4","first-page":"1","article-title":"A survey of dynamic neural networks","volume":"55","author":"Han","year":"2022","journal-title":"ACM Comput. Surv"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01261-8_25"},{"article-title":"Adaptive deep learning via dynamic inference paths","volume-title":"Proc. Eur. Conf. Comput. Vis. (ECCV)","author":"Teerapittayanon","key":"ref6"},{"key":"ref7","article-title":"Conditional computation in neural networks for faster models","author":"Bengio","year":"2015","journal-title":"arXiv preprint arXiv:1511.06297"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICPR.2016.7900006"},{"article-title":"McUNet: Memory-efficient inference of unrolled neural networks with early exits","volume-title":"Proc. NeurIPS","author":"Lin","key":"ref9"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3638242"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICCAR64901.2025.11072991"},{"issue":"3","key":"ref12","first-page":"1","article-title":"Dynamic neural networks in natural language processing: A survey","volume":"56","author":"Xu","year":"2024","journal-title":"ACM Comput. Surv"},{"key":"ref13","article-title":"Dynamic inference: A survey","author":"Huang","year":"2023","journal-title":"arXiv preprint arXiv:2306.04189"},{"article-title":"Improving early exits in deep networks via decision gates","volume-title":"Proc. NeurIPS","author":"Li","key":"ref14"},{"article-title":"Slimmable neural networks","volume-title":"Proc. Int. Conf. Learn. Representations (ICLR)","author":"Yu","key":"ref15"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00189"},{"key":"ref17","article-title":"Dynamic sparse training: Find efficient sparse network from scratch with trainable masked layers","author":"Liu","year":"2020","journal-title":"arXiv preprint arXiv:2005.06870"},{"article-title":"SplitEE: Split execution for efficient earlyexit inference on edge devices","volume-title":"Proc. ACM\/IEEE Int. Conf. Inf. Process. Sensor Netw. (IPSN)","author":"Li","key":"ref18"},{"article-title":"On calibration of modern neural networks","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Guo","key":"ref19"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICUAS57906.2023.10156376"},{"article-title":"Shallow-deep networks: Understanding and mitigating network overthinking","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Kaya","key":"ref21"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2022.3187002"},{"key":"ref23","first-page":"10336","article-title":"Dynamic inference with early exit in deep neural networks","volume-title":"Proc. Int. Conf. Pattern Recognit. (ICPR)","author":"Manessi"},{"article-title":"Dynamic channel gating for neural network acceleration","volume-title":"Proc. Int. Conf. Learn. Representations (ICLR)","author":"Gao","key":"ref24"},{"key":"ref25","first-page":"1","article-title":"Conditional early exit networks for energy-efficient object detection","volume-title":"Proc. IEEE\/ACM Int. Conf. Comput.-Aided Des. (ICCAD)","author":"Panda"},{"key":"ref26","first-page":"3693","article-title":"Edge-YOLO: An early-exit object detector for edge computing","volume-title":"Proc. IEEE Int. Conf. Acoustics, Speech and Signal Process. (ICASSP)","author":"Liu"},{"key":"ref27","first-page":"10574","article-title":"Energy-aware dynamic early-exit for mobile neural networks","volume-title":"Proc. AAAI Conf. Artif. Intell","author":"Zhang"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/MC.2022.3217841"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/DSN-W50199.2020.00014"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/MICRO50266.2020.00033"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ETS61313.2024.10568018"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/DFT52944.2021.9568281"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00721"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/FCCM.2016.22"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2015.7298594"}],"event":{"name":"2025 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)","start":{"date-parts":[[2025,7,6]]},"location":"Kalamata, Greece","end":{"date-parts":[[2025,7,9]]}},"container-title":["2025 IEEE Computer Society Annual Symposium on VLSI (ISVLSI)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11129697\/11130193\/11130246.pdf?arnumber=11130246","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,28]],"date-time":"2025-08-28T04:31:39Z","timestamp":1756355499000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11130246\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,7,6]]},"references-count":35,"URL":"https:\/\/doi.org\/10.1109\/isvlsi65124.2025.11130246","relation":{},"subject":[],"published":{"date-parts":[[2025,7,6]]}}}