{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T19:29:22Z","timestamp":1773170962373,"version":"3.50.1"},"reference-count":85,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"4","license":[{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,4,1]],"date-time":"2026-04-01T00:00:00Z","timestamp":1775001600000},"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":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,4]]},"DOI":"10.1109\/tpami.2025.3647880","type":"journal-article","created":{"date-parts":[[2025,12,24]],"date-time":"2025-12-24T18:45:42Z","timestamp":1766601942000},"page":"4896-4908","source":"Crossref","is-referenced-by-count":0,"title":["Toward Unified Expertise: Learning a Single Vision Model From Diverse Perception"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9424-4617","authenticated-orcid":false,"given":"Zitian","family":"Chen","sequence":"first","affiliation":[{"name":"College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-6556-8359","authenticated-orcid":false,"given":"Mingyu","family":"Ding","sequence":"additional","affiliation":[{"name":"Department of Electrical and Computer Engineering, University of California Berkeley, Berkeley, CA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yikang","family":"Shen","sequence":"additional","affiliation":[{"name":"IBM, Cambridge, MA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3778-9135","authenticated-orcid":false,"given":"Erik","family":"Learned-Miller","sequence":"additional","affiliation":[{"name":"College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4031-5886","authenticated-orcid":false,"given":"Chuang","family":"Gan","sequence":"additional","affiliation":[{"name":"College of Information and Computer Sciences, University of Massachusetts Amherst, Amherst, MA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01138"},{"key":"ref2","first-page":"2039","article-title":"Just pick a sign: Optimizing deep multitask models with gradient sign dropout","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Chen","year":"2020"},{"key":"ref3","first-page":"5824","article-title":"Gradient surgery for multi-task learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Yu","year":"2020"},{"key":"ref4","article-title":"UNIFIED-IO: A unified model for vision, language, and multi-modal tasks","volume-title":"Proc. 11th Int. Conf. Learn. Representations","author":"Lu","year":"2023"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01995"},{"key":"ref6","first-page":"4651","article-title":"Perceiver: General perception with iterative attention","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Jaegle","year":"2021"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-023-01931-6"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1023\/A:1007379606734"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.579"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00147"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.5244\/C.34.59"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00663"},{"key":"ref13","first-page":"8728","article-title":"AdaShare: Learning what to share for efficient deep multi-task learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Sun","year":"2020"},{"key":"ref14","article-title":"Routing networks: Adaptive selection of non-linear functions for multi-task learning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Rosenbaum","year":"2018"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19812-0_18"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1991.3.1.79"},{"key":"ref17","article-title":"Outrageously large neural networks: The sparsely-gated mixture-of-experts layer","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Shazeer","year":"2017"},{"key":"ref18","article-title":"GShard: Scaling giant models with conditional computation and automatic sharding","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Lepikhin","year":"2021"},{"issue":"120","key":"ref19","first-page":"1","article-title":"Switch transformers: Scaling to trillion parameter models with simple and efficient sparsity","volume-title":"J. Mach. Learn. Res.","volume":"23","author":"Fedus","year":"2022"},{"key":"ref20","first-page":"8583","article-title":"Scaling vision with sparse mixture of experts","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Riquelme","year":"2021"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-32010-6_300155"},{"key":"ref22","first-page":"9564","article-title":"Multimodal contrastive learning with limoe: The language-image mixture of experts","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Mustafa","year":"2022"},{"key":"ref23","first-page":"6265","article-title":"Base layers: Simplifying training of large, sparse models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Lewis","year":"2021"},{"key":"ref24","article-title":"Dense-to-sparse gate for mixture-of-experts","author":"Nie","year":"2021"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.emnlp-main.278"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.70"},{"key":"ref27","first-page":"28441","article-title":"M$^{3}$3ViT: Mixture-of-experts vision transformer for efficient multi-task learning with model-accelerator co-design","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liang","year":"2022"},{"key":"ref28","article-title":"Deep compression: Compressing deep neural networks with pruning, trained quantization and Huffman coding","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Han","year":"2016"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.24963\/ijcai.2018\/309"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00572"},{"key":"ref31","article-title":"Comparing rewinding and fine-tuning in neural network pruning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Renda","year":"2020"},{"key":"ref32","article-title":"Rethinking the value of network pruning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Liu","year":"2019"},{"key":"ref33","article-title":"Slimmable neural networks","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Yu","year":"2019"},{"issue":"2","key":"ref34","first-page":"3","article-title":"LoRA: Low-rank adaptation of large language models","volume":"1","author":"Hu","year":"2022","journal-title":"ICLR"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-023-01891-x"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00516"},{"key":"ref37","article-title":"Tip-Adapter: Training-free clip-adapter for better vision-language modeling","author":"Zhang","year":"2021"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20077-9_17"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-short.1"},{"key":"ref40","first-page":"16888","article-title":"Efficient test-time model adaptation without forgetting","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Niu","year":"2022"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01525"},{"key":"ref42","first-page":"38298","article-title":"Test-Time model adaptation with only forward passes","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Niu","year":"2024"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3117837"},{"key":"ref44","first-page":"527","article-title":"Adaptive neural networks for efficient inference","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Bolukbasi","year":"2017"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01246-5_1"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01261-8_25"},{"key":"ref47","article-title":"Multi-Scale dense networks for resource efficient image classification","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Huang","year":"2018"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58536-5_11"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00850"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00858"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11630"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1145\/3394171.3413915"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19803-8_30"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/WACV57701.2024.00769"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19809-0_29"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00873"},{"key":"ref57","first-page":"7482","article-title":"Multi-task learning using uncertainty to weigh losses for scene geometry and semantics","volume-title":"Proc. IEEE Conf. Comput. Vis. Pattern Recognit.","author":"Kendall","year":"2018"},{"key":"ref58","article-title":"Once for all: Train one network and specialize it for efficient deployment","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Cai","year":"2020"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.119"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00391"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.544"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00195"},{"key":"ref65","article-title":"Depth map prediction from a single image using a multi-scale deep network","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"27","author":"Eigen","year":"2014"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.324"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01228-1_26"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00197"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.433"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00332"},{"key":"ref71","article-title":"Large batch optimization for deep learning: Training bert in 76 minutes","author":"You","year":"2019"},{"key":"ref72","article-title":"An image is worth 16 x16 words: Transformers for image recognition at scale","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Dosovitskiy","year":"2021"},{"key":"ref73","article-title":"Decoupled weight decay regularization","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Loshchilov","year":"2019"},{"key":"ref74","article-title":"SGDR: Stochastic gradient descent with warm restarts","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Loshchilov","year":"2017"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20053-3_5"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2017.2723009"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00914"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2012.6248092"},{"key":"ref80","article-title":"The caltech-UCSD birds-200-2011 dataset","author":"Wah","year":"2011"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2013.77"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-009-0275-4"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.350"},{"key":"ref84","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-33715-4_54"},{"key":"ref85","first-page":"2790","article-title":"Parameter-efficient transfer learning for NLP","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Houlsby","year":"2019"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11424231\/11314543.pdf?arnumber=11314543","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,10]],"date-time":"2026-03-10T01:34:33Z","timestamp":1773106473000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11314543\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,4]]},"references-count":85,"journal-issue":{"issue":"4"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2025.3647880","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":[[2026,4]]}}}