{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,14]],"date-time":"2026-04-14T12:09:24Z","timestamp":1776168564551,"version":"3.50.1"},"reference-count":108,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"2","license":[{"start":{"date-parts":[[2024,2,1]],"date-time":"2024-02-01T00:00:00Z","timestamp":1706745600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,2,1]],"date-time":"2024-02-01T00:00:00Z","timestamp":1706745600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,2,1]],"date-time":"2024-02-01T00:00:00Z","timestamp":1706745600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/100020612","name":"Advanced Research and Technology Innovation Centre, College of Design and Engineering, National University of Singapore","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100020612","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001352","name":"National University of Singapore","doi-asserted-by":"publisher","award":["A-0005947-21-00"],"award-info":[{"award-number":["A-0005947-21-00"]}],"id":[{"id":"10.13039\/501100001352","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Singapore Ministry of Education Academic Research Fund Tier 1","award":["WBS: A-0009440-01-00"],"award-info":[{"award-number":["WBS: A-0009440-01-00"]}]},{"DOI":"10.13039\/501100001381","name":"National Research Foundation Singapore","doi-asserted-by":"publisher","award":["AISG2-RP-2021-023"],"award-info":[{"award-number":["AISG2-RP-2021-023"]}],"id":[{"id":"10.13039\/501100001381","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2024,2]]},"DOI":"10.1109\/tpami.2023.3329173","type":"journal-article","created":{"date-parts":[[2023,11,1]],"date-time":"2023-11-01T18:15:19Z","timestamp":1698862519000},"page":"896-912","source":"Crossref","is-referenced-by-count":228,"title":["MetaFormer Baselines for Vision"],"prefix":"10.1109","volume":"46","author":[{"ORCID":"https:\/\/orcid.org\/0000-0003-3349-5890","authenticated-orcid":false,"given":"Weihao","family":"Yu","sequence":"first","affiliation":[{"name":"National University of Singapore, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3354-1968","authenticated-orcid":false,"given":"Chenyang","family":"Si","sequence":"additional","affiliation":[{"name":"Sea AI Lab, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-3400-8943","authenticated-orcid":false,"given":"Pan","family":"Zhou","sequence":"additional","affiliation":[{"name":"Sea AI Lab, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8602-4681","authenticated-orcid":false,"given":"Mi","family":"Luo","sequence":"additional","affiliation":[{"name":"National University of Singapore, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0596-2087","authenticated-orcid":false,"given":"Yichen","family":"Zhou","sequence":"additional","affiliation":[{"name":"Sea AI Lab, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6843-0064","authenticated-orcid":false,"given":"Jiashi","family":"Feng","sequence":"additional","affiliation":[{"name":"Sea AI Lab, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8906-3777","authenticated-orcid":false,"given":"Shuicheng","family":"Yan","sequence":"additional","affiliation":[{"name":"Sea AI Lab, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0057-1404","authenticated-orcid":false,"given":"Xinchao","family":"Wang","sequence":"additional","affiliation":[{"name":"National University of Singapore, Singapore"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref2","article-title":"ResNet strikes back: An improved training procedure in timm","author":"Wightman","year":"2021"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52688.2022.01167"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00061"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.195"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-33266-1_8"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00474"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref10","first-page":"1691","article-title":"Generative pretraining from pixels","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Chen"},{"key":"ref11","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00681"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00060"},{"key":"ref15","first-page":"15908","article-title":"Transformer in transformer","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Han"},{"key":"ref16","article-title":"Refiner: Refining self-attention for vision transformers","author":"Zhou","year":"2021"},{"key":"ref17","first-page":"24261","article-title":"MLP-mixer: An all-MLP architecture for vision","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Tolstikhin"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.naacl-main.319"},{"key":"ref19","article-title":"A battle of network structures: An empirical study of CNN, transformer, and MLP","author":"Zhao","year":"2021"},{"key":"ref20","article-title":"On the connection between local attention and dynamic depth-wise convolution","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Han"},{"key":"ref21","first-page":"980","article-title":"Global filter networks for image classification","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Rao"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3206148"},{"key":"ref23","first-page":"10183","article-title":"Synthesizer: Rethinking self-attention for transformer models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tay"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52688.2022.01055"},{"key":"ref25","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Touvron"},{"key":"ref26","first-page":"9204","article-title":"Pay attention to MLPs","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liu"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00745"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1807.06521"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.5555\/3045118.3045167"},{"key":"ref30","article-title":"Layer normalization","author":"Ba","year":"2016"},{"key":"ref31","first-page":"1106","article-title":"ImageNet classification with deep convolutional neural networks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Krizhevsky"},{"key":"ref32","first-page":"3965","article-title":"CoAtNet: Marrying convolution and attention for all data sizes","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Dai"},{"key":"ref33","article-title":"NormFormer: Improved transformer pretraining with extra normalization","author":"Shleifer","year":"2021"},{"issue":"140","key":"ref34","first-page":"1","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"ref35","article-title":"PaLM: Scaling language modeling with pathways","author":"Chowdhery","year":"2022"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.5555\/3104322.3104425"},{"key":"ref37","article-title":"Improving language understanding by generative pre-training","author":"Radford","year":"2018"},{"key":"ref38","article-title":"Gaussian error linear units (GELUs)","author":"Hendrycks","year":"2016"},{"key":"ref39","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018"},{"key":"ref40","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Brown"},{"key":"ref41","article-title":"Primer: Searching for efficient transformers for language modeling","author":"So","year":"2021"},{"key":"ref42","article-title":"How many flops does tanh need?","year":"2017"},{"key":"ref43","first-page":"971","article-title":"Self-normalizing neural networks","volume-title":"Proc. Adv. Neural In. Process. Syst.","author":"Klambauer"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.123"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58529-7_21"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00010"},{"key":"ref47","first-page":"16410","article-title":"Gradinit: Learning to initialize neural networks for stable and efficient training","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zhu"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2020.emnlp-main.463"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00476"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01044"},{"key":"ref51","article-title":"Visual attention network","author":"Guo","year":"2022"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01181"},{"key":"ref53","article-title":"Dual vision transformer","author":"Yao","year":"2022"},{"key":"ref54","article-title":"Uniformer: Unified transformer for efficient spatiotemporal representation learning","author":"Li","year":"2022"},{"key":"ref55","article-title":"Inception transformer","author":"Si","year":"2022"},{"key":"ref56","first-page":"11963","article-title":"Scaling up your kernels to 31x31: Revisiting large kernel design in CNNs","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Ding"},{"key":"ref57","article-title":"More ConvNets in the 2020s: Scaling up kernels beyond 51x51 using sparsity","volume-title":"Proc. 11th Int. Conf. Learn. Representations","author":"Liu"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20053-3_27"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.5555\/3454287.3455008"},{"key":"ref62","article-title":"PyTorch image models","author":"Wightman","year":"2019"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW50498.2020.00359"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1007\/springerreference_178963"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00612"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i07.7000"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.308"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46493-0_39"},{"key":"ref69","article-title":"Multigrain: A unified image embedding for classes and instances","author":"Berman","year":"2019"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00815"},{"key":"ref71","article-title":"Adam: A method for stochastic optimization","author":"Kingma","year":"2014"},{"key":"ref72","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2017"},{"key":"ref73","article-title":"Large batch optimization for deep learning: Training bert in 76 minutes","author":"You","year":"2019"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1137\/0330046"},{"key":"ref75","article-title":"Benchmarking neural network robustness to common corruptions and perturbations","author":"Hendrycks","year":"2019"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01501"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00823"},{"key":"ref78","first-page":"10506","article-title":"Learning robust global representations by penalizing local predictive power","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wang"},{"key":"ref79","first-page":"10096","article-title":"EfficientNetV2: Smaller models and faster training","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan"},{"key":"ref80","first-page":"27378","article-title":"Understanding the robustness in vision transformers","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Zhou"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01173"},{"key":"ref82","article-title":"Can CNNs be more robust than transformers?","volume-title":"Proc. 11th Int. Conf. Learn. Representations","author":"Wang"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00644"},{"key":"ref86","article-title":"Longformer: The long-document transformer","author":"Beltagy","year":"2020"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00599"},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.544"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01228-1_26"},{"key":"ref90","article-title":"RoBERTa: A robustly optimized BERT pretraining approach","author":"Liu","year":"2019"},{"key":"ref91","article-title":"Tener: Adapting transformer encoder for named entity recognition","author":"Yan","year":"2019"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2020-3015"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.21437\/Interspeech.2019-2702"},{"key":"ref94","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33016706"},{"key":"ref95","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP40776.2020.9053591"},{"key":"ref96","doi-asserted-by":"publisher","DOI":"10.1145\/3394171.3413671"},{"key":"ref97","article-title":"VisualBERT: A simple and performant baseline for vision and language","author":"Li","year":"2019"},{"key":"ref98","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58577-8_7"},{"key":"ref99","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01058"},{"key":"ref100","first-page":"30008","article-title":"Focal attention for long-range interactions in vision transformers","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Yang"},{"key":"ref101","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3145427"},{"key":"ref102","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2023.3303397"},{"key":"ref103","article-title":"As-MLP: An axial shifted MLP architecture for vision","author":"Lian","year":"2021"},{"key":"ref104","article-title":"Focal modulation networks","author":"Yang","year":"2022"},{"key":"ref105","article-title":"Hornet: Efficient high-order spatial interactions with recursive gated convolutions","author":"Rao","year":"2022"},{"key":"ref106","article-title":"Pay less attention with lightweight and dynamic convolutions","author":"Wu","year":"2019"},{"key":"ref107","article-title":"Sequencer: Deep LSTM for image classification","author":"Tatsunami","year":"2022"},{"key":"ref108","article-title":"Vision GNN: An image is worth graph of nodes","author":"Han","year":"2022"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/34\/10384454\/10304335.pdf?arnumber=10304335","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,4,4]],"date-time":"2025-04-04T19:31:11Z","timestamp":1743795071000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10304335\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,2]]},"references-count":108,"journal-issue":{"issue":"2"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2023.3329173","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":[[2024,2]]}}}