{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,10]],"date-time":"2026-04-10T08:15:02Z","timestamp":1775808902536,"version":"3.50.1"},"reference-count":100,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"6","license":[{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key Research and Development Program of China","award":["2023YFF0713300"],"award-info":[{"award-number":["2023YFF0713300"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62231002"],"award-info":[{"award-number":["62231002"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Neural Netw. Learning Syst."],"published-print":{"date-parts":[[2025,6]]},"DOI":"10.1109\/tnnls.2025.3552654","type":"journal-article","created":{"date-parts":[[2025,4,16]],"date-time":"2025-04-16T13:51:46Z","timestamp":1744811506000},"page":"11617-11631","source":"Crossref","is-referenced-by-count":13,"title":["RefConv: Reparameterized Refocusing Convolution for Powerful ConvNets"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0009-0000-7986-6269","authenticated-orcid":false,"given":"Zhicheng","family":"Cai","sequence":"first","affiliation":[{"name":"School of Electronic Science and Engineering, Nanjing University, Nanjing, China"}]},{"given":"Xiaohan","family":"Ding","sequence":"additional","affiliation":[{"name":"Tencent AI Laboratory, Shenzhen, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3057-0608","authenticated-orcid":false,"given":"Qiu","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Electronic Science and Engineering, Nanjing University, Nanjing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3094-4371","authenticated-orcid":false,"given":"Xun","family":"Cao","sequence":"additional","affiliation":[{"name":"School of Electronic Science and Engineering, Nanjing University, Nanjing, China"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01167"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1807.06521"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00060"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00140"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01264-9_8"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00340"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v35i11.17226"},{"key":"ref11","first-page":"1","article-title":"Visualizing the loss landscape of neural nets","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Li"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00200"},{"key":"ref13","article-title":"Very deep convolutional networks for large-scale image recognition","author":"Simonyan","year":"2014","journal-title":"arXiv:1409.1556"},{"key":"ref14","article-title":"ResNeSt: Split-attention networks","author":"Zhang","year":"2020","journal-title":"arXiv:2004.08955"},{"key":"ref15","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017","journal-title":"arXiv:1704.04861"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01166"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01157"},{"key":"ref19","first-page":"10353","article-title":"HorNet: Efficient high-order spatial interactions with recursive gated convolutions","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Rao"},{"key":"ref20","article-title":"More ConvNets in the 2020s: Scaling up kernels beyond 51\u00d751 using sparsity","author":"Liu","year":"2022","journal-title":"arXiv:2207.03620"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00353"},{"key":"ref22","first-page":"7324","article-title":"Making convolutional networks shift-invariant again","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","volume":"97","author":"Zhang"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00572"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-43901-8_61"},{"key":"ref25","first-page":"1","article-title":"Gather-excite: Exploiting feature context in convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"31","author":"Hu"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01155"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00447"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01074"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01352"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00066"},{"key":"ref31","article-title":"MobileOne: An improved one millisecond mobile backbone","author":"Kumar Anasosalu Vasu","year":"2022","journal-title":"arXiv:2206.04040"},{"key":"ref32","article-title":"RepGhost: A hardware-efficient ghost module via re-parameterization","author":"Chen","year":"2022","journal-title":"arXiv:2211.06088"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/978-981-99-8079-6_29"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2209.02976"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00721"},{"key":"ref36","article-title":"PP-YOLOE: An evolved version of YOLO","author":"Xu","year":"2022","journal-title":"arXiv:2203.16250"},{"key":"ref37","first-page":"529","article-title":"Collapsible linear blocks for super-efficient super resolution","volume-title":"Proc. Mach. Learn. Syst.","volume":"4","author":"Bhardwaj"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW59228.2023.00189"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW56347.2022.00097"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA46639.2022.9812394"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1093\/bioinformatics\/btac120"},{"key":"ref42","article-title":"UniRepLKNet: A universal perception large-kernel ConvNet for audio, video, point cloud, time-series and image recognition","author":"Ding","year":"2023","journal-title":"arXiv:2311.15599"},{"key":"ref43","article-title":"DiracNets: Training very deep neural networks without skip-connections","author":"Zagoruyko","year":"2017","journal-title":"arXiv:1706.00388"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1111\/cgf.13340"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2022.3175432"},{"key":"ref46","first-page":"901","article-title":"Weight normalization: A simple reparameterization to accelerate training of deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"29","author":"Salimans"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.305"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11768"},{"key":"ref49","article-title":"DyNet: Dynamic convolution for accelerating convolutional neural networks","author":"Zhang","year":"2020","journal-title":"arXiv:2004.10694"},{"key":"ref50","article-title":"Omni-dimensional dynamic convolution","author":"Li","year":"2022","journal-title":"arXiv:2209.07947"},{"key":"ref51","article-title":"KernelWarehouse: Towards parameter-efficient dynamic convolution","author":"Li","year":"2023","journal-title":"arXiv:2308.08361"},{"key":"ref52","first-page":"1","article-title":"CondConv: Conditionally parameterized convolutions for efficient inference","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Yang"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01104"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58555-6_46"},{"key":"ref55","article-title":"HyperNetworks","author":"Ha","year":"2016","journal-title":"arXiv:1609.09106"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58577-8_6"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58523-5_41"},{"key":"ref58","first-page":"1","article-title":"Dynamic filter networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"29","author":"Brabandere"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1145\/3072959.3073708"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00265"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2024.3370294"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2023.3298263"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01544"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3066410"},{"key":"ref65","first-page":"1","article-title":"Pruning deep neural networks from a sparsity perspective","volume-title":"Proc. ICLR","author":"Diao"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3069886"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2021.3064293"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2017.2774288"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3275159"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3279281"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58580-8_40"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00341"},{"key":"ref74","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00062"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00063"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01186"},{"key":"ref79","first-page":"23495","article-title":"Inception transformer","volume-title":"Proc. Conf. Workshop Neural Inf. Process. Syst.","volume":"35","author":"Si"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3329173"},{"key":"ref81","first-page":"249","article-title":"Understanding the difficulty of training deep feedforward neural networks","volume-title":"Proc. 13th Int. Conf. Artif. Intell. Statist.","author":"Glorot"},{"key":"ref82","first-page":"1097","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst. (NIPS)","author":"Krizhevsky"},{"key":"ref83","article-title":"SqueezeNet: AlexNet-level accuracy with 50\u00d7 fewer parameters and <0.5MB model size","author":"Iandola","year":"2016","journal-title":"arXiv:1602.07360"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.634"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01044"},{"key":"ref86","first-page":"10096","article-title":"EfficientNetv2: Smaller models and faster training","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01506"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00532"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/ADICS58448.2024.10533619"},{"key":"ref90","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.660"},{"key":"ref91","article-title":"Rethinking Atrous convolution for semantic image segmentation","author":"Chen","year":"2017","journal-title":"arXiv:1706.05587"},{"key":"ref92","article-title":"Explaining and harnessing adversarial examples","author":"Goodfellow","year":"2014","journal-title":"arXiv:1412.6572"},{"key":"ref93","article-title":"Towards deep learning models resistant to adversarial attacks","author":"Madry","year":"2017","journal-title":"arXiv:1706.06083"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1109\/SP.2017.49"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00957"},{"key":"ref96","first-page":"1","article-title":"Learning both weights and connections for efficient neural network","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"28","author":"Han"},{"key":"ref97","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.12262"},{"key":"ref98","article-title":"Pruning filters for efficient ConvNets","author":"Li","year":"2016","journal-title":"arXiv:1608.08710"},{"key":"ref99","first-page":"1","article-title":"Dynamic network surgery for efficient DNNs","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"29","author":"Guo"},{"key":"ref100","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","author":"Han","year":"2015","journal-title":"arXiv:1510.00149"}],"container-title":["IEEE Transactions on Neural Networks and Learning Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/5962385\/11022714\/10966435.pdf?arnumber=10966435","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,5]],"date-time":"2025-12-05T18:39:36Z","timestamp":1764959976000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10966435\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6]]},"references-count":100,"journal-issue":{"issue":"6"},"URL":"https:\/\/doi.org\/10.1109\/tnnls.2025.3552654","relation":{},"ISSN":["2162-237X","2162-2388"],"issn-type":[{"value":"2162-237X","type":"print"},{"value":"2162-2388","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6]]}}}