{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T16:39:43Z","timestamp":1781887183115,"version":"3.54.5"},"reference-count":81,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62472222"],"award-info":[{"award-number":["62472222"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100004608","name":"Natural Science Foundation of Jiangsu Province","doi-asserted-by":"publisher","award":["BK20240080"],"award-info":[{"award-number":["BK20240080"]}],"id":[{"id":"10.13039\/501100004608","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Multimedia"],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/tmm.2025.3535405","type":"journal-article","created":{"date-parts":[[2025,1,27]],"date-time":"2025-01-27T13:54:42Z","timestamp":1737986082000},"page":"4670-4683","source":"Crossref","is-referenced-by-count":7,"title":["Prune and Merge: Efficient Token Compression for Vision Transformer With Spatial Information Preserved"],"prefix":"10.1109","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0009-0005-0831-9272","authenticated-orcid":false,"given":"Junzhu","family":"Mao","sequence":"first","affiliation":[{"name":"School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6344-9951","authenticated-orcid":false,"given":"Yang","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1956-3367","authenticated-orcid":false,"given":"Jinyang","family":"Guo","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Software Development Environment, Institute of Artificial Intelligence, Beihang University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0337-9410","authenticated-orcid":false,"given":"Yazhou","family":"Yao","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Nanjing University of Science and Technology, Nanjing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8232-5049","authenticated-orcid":false,"given":"Xiansheng","family":"Hua","sequence":"additional","affiliation":[{"name":"Terminus Group, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2999-2088","authenticated-orcid":false,"given":"Hengtao","family":"Shen","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, University of Electronic Science and Technology of China, Chengdu, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3547806"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01998"},{"key":"ref3","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Dosovitskiy","year":"2021"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2021.3120873"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3244340"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73383-3_13"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/TNNLS.2022.3202571"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3251100"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3264883"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3275308"},{"key":"ref12","article-title":"Deformable DETR: Deformable transformers for end-to-end object detection","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zhu","year":"2021"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02617"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00681"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01075"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2023.3249579"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72946-1_25"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02145"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00756"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00033"},{"key":"ref22","first-page":"10347","article-title":"Training data-efficient image transformers & distillation through attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Touvron","year":"2021"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00060"},{"key":"ref24","first-page":"31333","article-title":"A unified sequence interface for vision tasks","volume-title":"Proc. Adv. Neural Inform. Process. Syst.","author":"Chen","year":"2022"},{"key":"ref25","article-title":"Unified-IO: A unified model for vision, language, and multi-modal tasks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Lu","year":"2022"},{"key":"ref26","article-title":"Reducing transformer depth on demand with structured dropout","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Fan","year":"2020"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00413"},{"key":"ref28","article-title":"Reweighted proximal pruning for large-scale language representation","author":"Guo","year":"2019"},{"key":"ref29","first-page":"14037","article-title":"Are sixteen heads really better than one?","volume-title":"Proc. 33rd Int. Conf. Neural Inf. Process. Syst.","author":"Michel","year":"2019"},{"key":"ref30","article-title":"Structured pruning of a bert-based question answering model","author":"McCarley","year":"2019"},{"key":"ref31","first-page":"15834","article-title":"The lottery ticket hypothesis for pre-trained bert networks","volume-title":"Proc. Adv. Neural Inform. Process. Syst.","author":"Chen","year":"2020"},{"key":"ref32","first-page":"19974","article-title":"Chasing sparsity in vision transformers: An end-to-end exploration","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Chen","year":"2021"},{"key":"ref33","article-title":"Unified visual transformer compression","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Yu","year":"2021"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-022-3646-6"},{"key":"ref35","article-title":"Visual transformer pruning","author":"Zhu","year":"2021"},{"key":"ref36","first-page":"3690","article-title":"Power-BERT: Accelerating bert inference via progressive word-vector elimination","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Goyal","year":"2020"},{"key":"ref37","first-page":"24898","article-title":"Ia-red $^{2}$: Interpretability-aware redundancy reduction for vision transformers","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Pan","year":"2021"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00043"},{"key":"ref39","first-page":"13937","article-title":"DynamicViT: Efficient vision transformers with dynamic token sparsification","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Rao","year":"2021"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01185"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v36i3.20202"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20083-0_37"},{"key":"ref43","article-title":"EViT: Expediting vision transformers via token reorganizations","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Liang","year":"2022"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20083-0_26"},{"key":"ref45","article-title":"Token merging: Your ViT but faster","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Bolya","year":"2023"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00600"},{"key":"ref47","first-page":"35462","article-title":"Expediting large-scale vision transformer for dense prediction without fine-tuning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liang","year":"2022"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2023.3293763"},{"key":"ref49","first-page":"598","article-title":"Optimal brain damage","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"LeCun","year":"1990"},{"key":"ref50","first-page":"2498","article-title":"Variational dropout sparsifies deep neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Molchanov","year":"2017"},{"key":"ref51","article-title":"DSD: Dense-sparse-dense training for deep neural networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Han","year":"2016"},{"key":"ref52","first-page":"20852","article-title":"The generalization-stability tradeoff in neural network pruning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Bartoldson","year":"2020"},{"key":"ref53","article-title":"On large-batch training for deep learning: Generalization gap and sharp minima","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Keskar","year":"2016"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.01204"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00061"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00675"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00476"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3265159"},{"key":"ref59","first-page":"9010","article-title":"SAVit: Structure-aware vision transformer pruning via collaborative optimization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Zheng","year":"2022"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2024.3355890"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW60793.2023.00085"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.721"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00208"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.74"},{"key":"ref65","article-title":"Pruning convolutional neural networks for resource efficient inference","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Molchanov","year":"2017"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01152"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/WACV57701.2024.00141"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01521"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00436"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-018-1140-0"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"ref74","article-title":"Decoupled weight decay regularization","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Loshchilov","year":"2018"},{"key":"ref75","article-title":"How to train your ViT? Data, augmentation, and regularization in vision transformers","author":"Steiner","year":"2022","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref76","first-page":"6105","article-title":"EfficientNet: Rethinking model scaling for convolutional neural networks","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Tan","year":"2019"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01170"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01181"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/iccv48922.2021.00717"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.5244\/c.35.204"},{"key":"ref81","first-page":"7793","article-title":"SoftSort: A continuous relaxation for the argsort operator","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Prillo","year":"2020"}],"container-title":["IEEE Transactions on Multimedia"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6046\/10844992\/10855481.pdf?arnumber=10855481","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,1]],"date-time":"2025-08-01T18:15:37Z","timestamp":1754072137000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10855481\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":81,"URL":"https:\/\/doi.org\/10.1109\/tmm.2025.3535405","relation":{},"ISSN":["1520-9210","1941-0077"],"issn-type":[{"value":"1520-9210","type":"print"},{"value":"1941-0077","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025]]}}}