{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,5]],"date-time":"2026-06-05T15:56:39Z","timestamp":1780674999338,"version":"3.54.1"},"reference-count":88,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"8","license":[{"start":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T00:00:00Z","timestamp":1754006400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T00:00:00Z","timestamp":1754006400000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T00:00:00Z","timestamp":1754006400000},"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":["2022ZD0160101"],"award-info":[{"award-number":["2022ZD0160101"]}]},{"DOI":"10.13039\/100007219","name":"Natural Science Foundation of Shanghai Municipality","doi-asserted-by":"publisher","award":["23ZR1402900"],"award-info":[{"award-number":["23ZR1402900"]}],"id":[{"id":"10.13039\/100007219","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Shanghai Science and Technology Commission Explorer Program Project","award":["24TS1401300"],"award-info":[{"award-number":["24TS1401300"]}]},{"name":"Shanghai Municipal Science and Technology Major Project","award":["2021SHZDZX0103"],"award-info":[{"award-number":["2021SHZDZX0103"]}]},{"name":"JC STEM Lab of AI for Science and Engineering"},{"name":"The Hong Kong Jockey Club Charities Trust"},{"name":"Research Grants Council of Hong Kong","award":["CUHK14213224"],"award-info":[{"award-number":["CUHK14213224"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2025,8]]},"DOI":"10.1109\/tpami.2025.3558421","type":"journal-article","created":{"date-parts":[[2025,4,7]],"date-time":"2025-04-07T21:44:35Z","timestamp":1744062275000},"page":"6463-6479","source":"Crossref","is-referenced-by-count":2,"title":["Stimulative Training++: Go Beyond the Performance Limits of Residual Networks"],"prefix":"10.1109","volume":"47","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-6272-2792","authenticated-orcid":false,"given":"Peng","family":"Ye","sequence":"first","affiliation":[{"name":"School of Information Science and Technology, Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2772-9320","authenticated-orcid":false,"given":"Tong","family":"He","sequence":"additional","affiliation":[{"name":"Shanghai AI Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shengji","family":"Tang","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9032-3991","authenticated-orcid":false,"given":"Baopu","family":"Li","sequence":"additional","affiliation":[{"name":"Baidu, Sunnyvale, CA, USA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0779-9818","authenticated-orcid":false,"given":"Tao","family":"Chen","sequence":"additional","affiliation":[{"name":"School of Information Science and Technology, Fudan University, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3378-7201","authenticated-orcid":false,"given":"Lei","family":"Bai","sequence":"additional","affiliation":[{"name":"Shanghai AI Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9163-2761","authenticated-orcid":false,"given":"Wanli","family":"Ouyang","sequence":"additional","affiliation":[{"name":"Shanghai AI Laboratory, Shanghai, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_38"},{"key":"ref3","first-page":"342","article-title":"The shattered gradients problem: If resnets are the answer, then what is the question","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Balduzzi"},{"key":"ref4","first-page":"550","article-title":"Residual networks behave like ensembles of relatively shallow networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Veit"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1016\/j.neunet.2022.01.012"},{"key":"ref6","first-page":"1","article-title":"A kernel perspective of skip connections in convolutional networks","volume-title":"Proc. 11th Int. Conf. Learn. Representations","author":"Barzilai"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1016\/0022-1031(74)90033-X"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1177\/014616727700300406"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1037\/0022-3514.61.4.570"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1111\/j.2044-8325.1995.tb00591.x"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1037\/0022-3514.37.6.822"},{"issue":"1","key":"ref12","first-page":"58","article-title":"Social loafing: A review of the literature","volume":"15","author":"Simms","year":"2014","journal-title":"J. Manage. Policy Pract."},{"key":"ref13","article-title":"UniformAugment: A search-free probabilistic data augmentation approach","author":"LingChen","year":"2020"},{"key":"ref14","first-page":"3596","article-title":"Stimulative training of residual networks: A social psychology perspective of loafing","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ye"},{"key":"ref15","article-title":"Distilling the knowledge in a neural network","author":"Hinton","year":"2015"},{"key":"ref16","first-page":"1","article-title":"Better teacher better student: Dynamic prior knowledge for knowledge distillation","volume-title":"Proc. 11th Int. Conf. Learn. Representations","author":"Zong"},{"key":"ref17","first-page":"2765","article-title":"Paraphrasing complex network: Network compression via factor transfer","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Kim"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5963"},{"key":"ref19","article-title":"One-for-all: Bridge the gap between heterogeneous architectures in knowledge distillation","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Hao"},{"key":"ref20","article-title":"FitNets: Hints for thin deep nets","author":"Romero","year":"2014"},{"key":"ref21","article-title":"Like what you like: Knowledge distill via neuron selectivity transfer","author":"Huang","year":"2017"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33013779"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00077"},{"key":"ref24","article-title":"ViTKD: Practical guidelines for ViT feature knowledge distillation","author":"Yang","year":"2022"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00454"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00381"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01060"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00037"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-022-01663-z"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01385"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01060"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3548149"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00536"},{"key":"ref34","first-page":"1","article-title":"Once-for-all: Train one network and specialize it for efficient deployment","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Cai"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58571-6_41"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_39"},{"key":"ref37","first-page":"84","article-title":"ImageNet classification with deep convolutional","volume":"60","author":"Krizhevsky","year":"2017","journal-title":"Neural Netw."},{"key":"ref38","first-page":"1","article-title":"An image is worth 16 \u00d7 16 words: Transformers for image recognition at scale","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Dosovitskiy"},{"key":"ref39","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan"},{"key":"ref40","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Touvron"},{"key":"ref41","article-title":"Improved regularization of convolutional neural networks with cutout","author":"DeVries","year":"2017"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4899-7502-7_79-1"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00612"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i07.7000"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW50498.2020.00359"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00020"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00081"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1605.07648"},{"key":"ref49","first-page":"10750","article-title":"DropBlock: A regularization method for convolutional networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ghiasi"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2019.2960566"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00396"},{"key":"ref52","article-title":"Are we done with ImageNet","author":"Beyer","year":"2020"},{"key":"ref53","first-page":"18661","article-title":"Supervised contrastive learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Khosla"},{"key":"ref54","first-page":"1139","article-title":"On the importance of initialization and momentum in deep learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Sutskever"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref56","first-page":"1","article-title":"Fixing weight decay regularization in Adam","volume-title":"Proc. 6th Int. Conf. Learn. Representations","author":"Loshchilov","year":"2018"},{"key":"ref57","first-page":"1","article-title":"Large batch optimization for deep learning: Training BERT in 76 minutes","volume-title":"Proc. Int. Conf. Learn. Representations","author":"You"},{"key":"ref58","first-page":"876","article-title":"Averaging weights leads to wider optima and better generalization","volume-title":"Proc. 34th Conf. Uncertainty Artif. Intell.","author":"Izmailov"},{"key":"ref59","article-title":"MultiGrain: A unified image embedding for classes and instances","author":"Berman","year":"2019"},{"key":"ref60","first-page":"8250","article-title":"Fixing the train-test resolution discrepancy","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Touvron"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00065"},{"key":"ref62","first-page":"1","article-title":"ResNet strikes back: An improved training procedure in TIMM","volume-title":"Proc. NeurIPS 2021 Workshop ImageNet: Past, Present, Future","author":"Wightman"},{"key":"ref63","article-title":"How to train state-of-the-art models using torchvision\u2019s latest primitives","author":"Vryniotis","year":"2023","journal-title":"PyTorch Blog"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20053-3_30"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00140"},{"key":"ref66","first-page":"6572","article-title":"Neural ordinary differential equations","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Chen"},{"key":"ref67","first-page":"23631","article-title":"Mitigating neural network overconfidence with logit normalization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Wei"},{"key":"ref68","first-page":"4898","article-title":"Understanding the effective receptive field in deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Luo"},{"key":"ref69","article-title":"MobileNets: Efficient convolutional neural networks for mobile vision applications","author":"Howard","year":"2017"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.243"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00716"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01166"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00061"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref75","first-page":"14207","article-title":"GradAug: A new regularization method for deep neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Yang"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00098"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.169"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00644"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00972"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00978"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00656"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01231-1_29"},{"key":"ref84","article-title":"MMDetection: Open MMLab detection toolbox and benchmark","author":"Chen","year":"2019"},{"key":"ref85","article-title":"MMSegmentation: OpenMMLab semantic segmentation toolbox and benchmark","author":"Contributors","year":"2020"},{"key":"ref86","article-title":"OpenMMLab pose estimation toolbox and benchmark","year":"2020"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01352"},{"key":"ref88","article-title":"Polynomial regression as an alternative to neural nets","author":"Cheng","year":"2018"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11068886\/10950128.pdf?arnumber=10950128","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,7,8]],"date-time":"2025-07-08T17:38:19Z","timestamp":1751996299000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10950128\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,8]]},"references-count":88,"journal-issue":{"issue":"8"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2025.3558421","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,8]]}}}