{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,10]],"date-time":"2026-02-10T16:07:44Z","timestamp":1770739664684,"version":"3.49.0"},"reference-count":82,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"3","license":[{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,3,1]],"date-time":"2026-03-01T00:00:00Z","timestamp":1772323200000},"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":["92470118"],"award-info":[{"award-number":["92470118"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62306176"],"award-info":[{"award-number":["62306176"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62550004"],"award-info":[{"award-number":["62550004"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100007219","name":"Natural Science Foundation of Shanghai","doi-asserted-by":"publisher","award":["23ZR1428700J"],"award-info":[{"award-number":["23ZR1428700J"]}],"id":[{"id":"10.13039\/100007219","id-type":"DOI","asserted-by":"publisher"}]},{"name":"XPlorer Prize"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,3]]},"DOI":"10.1109\/tpami.2025.3625728","type":"journal-article","created":{"date-parts":[[2025,10,27]],"date-time":"2025-10-27T17:57:52Z","timestamp":1761587872000},"page":"2445-2457","source":"Crossref","is-referenced-by-count":0,"title":["Neural Eigenfunctions are Structured Representation Learners"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-0932-1631","authenticated-orcid":false,"given":"Zhijie","family":"Deng","sequence":"first","affiliation":[{"name":"School of Computer Science, Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiaxin","family":"Shi","sequence":"additional","affiliation":[{"name":"Google DeepMind, London, U.K."}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Halıcıoğlu Data Science Institute and the Department of Computer Science and Engineering, University of California, San Diego, CA, USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-4216-1912","authenticated-orcid":false,"given":"Peng","family":"Cui","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, Tsinghua University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1533-8576","authenticated-orcid":false,"given":"Cewu","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Artificial Intelligence, Shanghai Jiao Tong University, Shanghai, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6254-2388","authenticated-orcid":false,"given":"Jun","family":"Zhu","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, Tsinghua University, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"EigenGame: PCA as a Nash equilibrium","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Gemp"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1561\/2200000002"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1016\/B978-012099975-0.50005-1"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1162\/089976603321780317"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1126\/science.290.5500.2323"},{"key":"ref6","article-title":"Representation learning with contrastive predictive coding","author":"Oord","year":"2018"},{"key":"ref7","article-title":"Improving language understanding by generative pre-training","author":"Radford","year":"2018"},{"key":"ref8","first-page":"9912","article-title":"Unsupervised learning of visual features by contrasting cluster assignments","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Caron"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.5555\/3524938.3525087"},{"key":"ref10","first-page":"5000","article-title":"Provable guarantees for self-supervised deep learning with spectral contrastive loss","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"HaoChen"},{"key":"ref11","article-title":"Contrastive and non-contrastive self-supervised learning recover global and local spectral embedding methods","author":"Balestriero","year":"2022"},{"key":"ref12","article-title":"Contrastive learning can find an optimal basis for approximately invariant functions","volume-title":"Proc. 1st Workshop Pre-Training: Perspectives Pitfalls Paths Forward ICML","author":"Johnson"},{"key":"ref13","first-page":"12310","article-title":"Barlow twins: Self-supervised learning via redundancy reduction","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Zbontar"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref15","first-page":"22118","article-title":"Open graph benchmark: Datasets for machine learning on graphs","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Hu"},{"key":"ref16","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2016"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/34.868688"},{"key":"ref18","article-title":"Spectral inference networks: Unifying deep and spectral learning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Pfau"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.5555\/3045118.3045167"},{"key":"ref20","first-page":"1","article-title":"Automatic differentiation in machine learning: A survey","volume":"18","author":"Baydin","year":"2018","journal-title":"J. Mach. Learn. Res."},{"key":"ref21","article-title":"Contrastive learning can find an optimal basis for approximately view-invariant functions","volume-title":"Proc. 11th Int. Conf. Learn. Representations","author":"Johnson"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/MSP.2012.2211477"},{"key":"ref23","first-page":"30233","article-title":"Matryoshka representation learning","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Kusupati"},{"key":"ref24","article-title":"Eigenvalues of the Laplacian and their relationship to the connectedness of a graph","author":"Marsden","year":"2013"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1145\/2939672.2939754"},{"key":"ref26","first-page":"1025","article-title":"Inductive representation learning on large graphs","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Hamilton"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1007\/BF02547521"},{"key":"ref28","first-page":"682","article-title":"Using the Nystr\u00f6m method to speed up kernel machines","volume-title":"Proc. 14th Annu. Conf. Neural Inf. Process. Syst.","author":"Williams"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1162\/0899766041732396"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.167"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.320"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46466-4_5"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46487-9_40"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.638"},{"key":"ref35","article-title":"Unsupervised representation learning by predicting image rotations","author":"Gidaris"},{"key":"ref36","first-page":"737","article-title":"Signature verification using a \u201cSiames","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Bromley"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2006.100"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00393"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00975"},{"key":"ref40","article-title":"Pushing the limits of self-supervised ResNets: Can we outperform supervised learning without labels on ImageNet?","author":"Tomasev","year":"2022"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.5555\/3495724.3497510"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01549"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00951"},{"key":"ref44","first-page":"8799","article-title":"VICRegL: Self-supervised learning of local visual features","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Bardes"},{"key":"ref45","article-title":"On the duality between contrastive and non-contrastive self-supervised learning","author":"Garrido","year":"2022"},{"key":"ref46","article-title":"VICReg: Variance-invariance-covariance regularization for self-supervised learning","author":"Bardes","year":"2021"},{"key":"ref47","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Radford"},{"key":"ref48","article-title":"How much can clip benefit vision-and-language tasks?","author":"Shen","year":"2021"},{"key":"ref49","article-title":"Evaluating clip: Towards characterization of broader capabilities and downstream implications","author":"Agarwal","year":"2021"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19815-1_40"},{"key":"ref51","article-title":"BEiT: BERT pre-training of image transformers","author":"Bao","year":"2021"},{"key":"ref52","article-title":"iBOT: Image BERT pre-training with online tokenizer","author":"Zhou","year":"2021"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01553"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19821-2_26"},{"key":"ref55","article-title":"Mugs: A multi-granular self-supervised learning framework","author":"Zhou","year":"2022"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01855"},{"key":"ref57","article-title":"Deep contrastive learning is provably (almost) principal component analysis","author":"Tian","year":"2022"},{"key":"ref58","article-title":"A theoretical analysis of contrastive unsupervised representation learning","author":"Arora","year":"2019"},{"key":"ref59","article-title":"For self-supervised learning, rationality implies generalization, provably","author":"Bansal","year":"2020"},{"key":"ref60","first-page":"309","article-title":"Predicting what you already know helps: Provable self-supervised learning","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Lee"},{"key":"ref61","article-title":"Understanding self-supervised learning with dual deep networks","author":"Tian","year":"2020"},{"key":"ref62","first-page":"1179","article-title":"Contrastive learning, multi-view redundancy, and linear models","volume-title":"Proc. Int. Conf. Algorithmic Learn. Theory","author":"Tosh"},{"key":"ref63","article-title":"Self-supervised learning from a multi-view perspective","author":"Tsai","year":"2020"},{"key":"ref64","first-page":"9929","article-title":"Understanding contrastive representation learning through alignment and uniformity on the hypersphere","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Wang"},{"key":"ref65","article-title":"PaLM: Scaling language modeling with pathways","author":"Chowdhery","year":"2022"},{"key":"ref66","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. Int. Conf. Neural Inf. Process. Syst.","author":"Brown"},{"key":"ref67","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1706.03762"},{"key":"ref69","first-page":"1691","article-title":"Generative pretraining from pixels","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Chen"},{"key":"ref70","article-title":"An image is worth 16 x 16 words: Transformers for image recognition at scale","author":"Dosovitskiy","year":"2020"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1145\/1390156.1390294"},{"key":"ref72","article-title":"OPT: Open pre-trained transformer language models","author":"Zhang","year":"2022"},{"key":"ref73","first-page":"1","article-title":"Adam: A method for stochastic optimization","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Kingma"},{"key":"ref74","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1145\/1646396.1646452"},{"key":"ref76","article-title":"The Pascal visual object classes challenge 2012 (VOC2012) results","author":"Everingham","year":"2012"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1145\/1460096.1460104"},{"key":"ref78","article-title":"Improved baselines with momentum contrastive learning","author":"Chen","year":"2020"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.322"},{"key":"ref81","article-title":"Detectron","author":"Girshick","year":"2018"},{"key":"ref82","article-title":"Large batch training of convolutional networks","author":"You","year":"2017"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11372200\/11218204.pdf?arnumber=11218204","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,2,9]],"date-time":"2026-02-09T21:05:33Z","timestamp":1770671133000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11218204\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3]]},"references-count":82,"journal-issue":{"issue":"3"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2025.3625728","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":[[2026,3]]}}}