{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T18:12:38Z","timestamp":1780596758254,"version":"3.54.1"},"reference-count":41,"publisher":"Elsevier BV","license":[{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/tdm\/userlicense\/1.0\/"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/www.elsevier.com\/legal\/tdmrep-license"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-017"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-012"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,11,1]],"date-time":"2026-11-01T00:00:00Z","timestamp":1793491200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-004"}],"content-domain":{"domain":["elsevier.com","sciencedirect.com"],"crossmark-restriction":true},"short-container-title":["Neural Networks"],"published-print":{"date-parts":[[2026,11]]},"DOI":"10.1016\/j.neunet.2026.109172","type":"journal-article","created":{"date-parts":[[2026,5,29]],"date-time":"2026-05-29T15:55:59Z","timestamp":1780070159000},"page":"109172","update-policy":"https:\/\/doi.org\/10.1016\/elsevier_cm_policy","source":"Crossref","is-referenced-by-count":0,"special_numbering":"C","title":["When task performance deceives: Task-geometry decoupling in learnable-curvature hyperbolic GNNs"],"prefix":"10.1016","volume":"203","author":[{"ORCID":"https:\/\/orcid.org\/0009-0004-1320-4165","authenticated-orcid":false,"given":"Lixian","family":"Chen","sequence":"first","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingchao","family":"Wang","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0008-5479-2580","authenticated-orcid":false,"given":"Zhaorong","family":"Dai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hanqian","family":"Liu","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-5501-5619","authenticated-orcid":false,"given":"Danxiang","family":"Ai","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-3928-7495","authenticated-orcid":false,"given":"Yang","family":"Shi","sequence":"additional","affiliation":[],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"78","reference":[{"key":"10.1016\/j.neunet.2026.109172_bib0001","series-title":"International conference on machine learning","first-page":"486","article-title":"Constant curvature graph convolutional networks","author":"Bachmann","year":"2020"},{"issue":"Nov","key":"10.1016\/j.neunet.2026.109172_bib0002","first-page":"463","article-title":"Rademacher and gaussian complexities: Risk bounds and structural results","volume":"3","author":"Bartlett","year":"2002","journal-title":"Journal of Machine Learning Research"},{"key":"10.1016\/j.neunet.2026.109172_bib0003","unstructured":"Bdeir, A., & Landwehr, N. (2024). Optimizing curvature learning for robust hyperbolic deep learning in computer vision. arXiv e-prints, (pp. 2405)."},{"key":"10.1016\/j.neunet.2026.109172_bib0004","unstructured":"B\u00e9cigneul, G., & Ganea, O.-E. (2018). Riemannian adaptive optimization methods. arXiv: 1810.00760."},{"issue":"4","key":"10.1016\/j.neunet.2026.109172_bib0005","doi-asserted-by":"crossref","first-page":"18","DOI":"10.1109\/MSP.2017.2693418","article-title":"Geometric deep learning: Going beyond euclidean data","volume":"34","author":"Bronstein","year":"2017","journal-title":"IEEE Signal Processing Magazine"},{"issue":"59\u2013115","key":"10.1016\/j.neunet.2026.109172_bib0006","first-page":"2","article-title":"Hyperbolic geometry","volume":"31","author":"Cannon","year":"1997","journal-title":"Flavors of Geometry"},{"key":"10.1016\/j.neunet.2026.109172_bib0007","doi-asserted-by":"crossref","DOI":"10.1007\/978-1-4757-2201-7","article-title":"Riemannian geometry","author":"do Carmo","year":"1992"},{"key":"10.1016\/j.neunet.2026.109172_bib0008","doi-asserted-by":"crossref","DOI":"10.1016\/j.automatica.2023.111466","article-title":"A control theoretic framework for adaptive gradient optimizers","volume":"160","author":"Chakrabarti","year":"2024","journal-title":"Automatica"},{"key":"10.1016\/j.neunet.2026.109172_bib0009","first-page":"4869","article-title":"Hyperbolic graph convolutional neural networks","volume":"32","author":"Chami","year":"2019","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109172_bib0010","first-page":"4891","article-title":"Measures of distortion for machine learning","volume":"31","author":"Chennuru Vankadara","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"3","key":"10.1016\/j.neunet.2026.109172_bib0011","doi-asserted-by":"crossref","first-page":"290","DOI":"10.1016\/0022-1236(67)90017-1","article-title":"The sizes of compact subsets of hilbert space and continuity of gaussian processes","volume":"1","author":"Dudley","year":"1967","journal-title":"Journal of Functional Analysis"},{"key":"10.1016\/j.neunet.2026.109172_bib0012","series-title":"2021\u202fIEEE International conference on data mining (ICDM)","first-page":"111","article-title":"ACE-HGNN: Adaptive curvature exploration hyperbolic graph neural network","author":"Fu","year":"2021"},{"key":"10.1016\/j.neunet.2026.109172_bib0013","first-page":"5345","article-title":"Hyperbolic neural networks","volume":"31","author":"Ganea","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109172_bib0014","series-title":"International conference on learning representations","article-title":"Learning mixed-curvature representations in product spaces","author":"Gu","year":"2018"},{"key":"10.1016\/j.neunet.2026.109172_bib0015","unstructured":"Gulcehre, C., Denil, M., Malinowski, M., Razavi, A., Pascanu, R., Hermann, K. M., Battaglia, P., Bapst, V., Raposo, D., Santoro, A. et al. (2018). Hyperbolic attention networks. arXiv: 1805.09786."},{"key":"10.1016\/j.neunet.2026.109172_bib0016","series-title":"Proceedings of the AAAI conference on artificial intelligence","first-page":"11754","article-title":"GraphmoRE: Mitigating topological heterogeneity via mixture of riemannian experts","volume":"vol. 39","author":"Guo","year":"2025"},{"key":"10.1016\/j.neunet.2026.109172_bib0017","unstructured":"Hu, W., Fey, M., Zitnik, M., Dong, Y., Ren, H., Liu, B., Catasta, M., & Leskovec, J. (2020). Open graph benchmark: Datasets for machine learning on graphs. arXiv: 2005.00687."},{"key":"10.1016\/j.neunet.2026.109172_bib0018","doi-asserted-by":"crossref","unstructured":"Kasarla, T., van Spengler, M., & Mettes, P. (2025). Balanced hyperbolic embeddings are natural out-of-distribution detectors. arXiv: 2506.10146.","DOI":"10.1007\/s11263-026-02775-6"},{"key":"10.1016\/j.neunet.2026.109172_bib0019","unstructured":"Katsman, I., & Gilbert, A. (2024). Shedding light on problems with hyperbolic graph learning. arXiv: 2411.06688."},{"key":"10.1016\/j.neunet.2026.109172_bib0020","series-title":"Proceedings of the IEEE\/CVF conference on computer vision and pattern recognition","first-page":"6418","article-title":"Hyperbolic image embeddings","author":"Khrulkov","year":"2020"},{"key":"10.1016\/j.neunet.2026.109172_bib0021","series-title":"International conference on learning representations (ICLR)","article-title":"Semi-supervised classification with graph convolutional networks","author":"Kipf","year":"2017"},{"key":"10.1016\/j.neunet.2026.109172_bib0022","article-title":"Probability in banach spaces: isoperimetry and processes","author":"Ledoux","year":"2013"},{"key":"10.1016\/j.neunet.2026.109172_bib0023","series-title":"The first learning on graphs conference","article-title":"On the expressiveness and generalization of hypergraph neural networks","author":"Luo","year":"2023"},{"key":"10.1016\/j.neunet.2026.109172_bib0024","series-title":"International conference on machine learning","first-page":"24925","article-title":"The numerical stability of hyperbolic representation learning","author":"Mishne","year":"2023"},{"key":"10.1016\/j.neunet.2026.109172_bib0025","series-title":"Foundations of Machine Learning","author":"Mohri","year":"2018"},{"issue":"8","key":"10.1016\/j.neunet.2026.109172_bib0026","doi-asserted-by":"crossref","DOI":"10.1088\/1742-5468\/adf296","article-title":"Optimal protocols for continual learning via statistical physics and control theory","volume":"2025","author":"Mori","year":"2025","journal-title":"Journal of Statistical Mechanics: Theory and Experiment"},{"key":"10.1016\/j.neunet.2026.109172_bib0027","series-title":"Conference on learning theory","first-page":"1376","article-title":"Norm-based capacity control in neural networks","author":"Neyshabur","year":"2015"},{"key":"10.1016\/j.neunet.2026.109172_bib0028","first-page":"6338","article-title":"Poincar\u00e9 embeddings for learning hierarchical representations","volume":"30","author":"Nickel","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109172_bib0029","series-title":"International conference on machine learning","first-page":"3779","article-title":"Learning continuous hierarchies in the lorentz model of hyperbolic geometry","author":"Nickel","year":"2018"},{"key":"10.1016\/j.neunet.2026.109172_bib0030","first-page":"20779","article-title":"Curvature regularization to prevent distortion in graph embedding","volume":"33","author":"Pei","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109172_bib0031","series-title":"International conference on machine learning","first-page":"4460","article-title":"Representation tradeoffs for hyperbolic embeddings","author":"Sala","year":"2018"},{"key":"10.1016\/j.neunet.2026.109172_bib0032","series-title":"Understanding machine learning: From theory to algorithms","author":"Shalev-Shwartz","year":"2014"},{"key":"10.1016\/j.neunet.2026.109172_bib0033","doi-asserted-by":"crossref","first-page":"91220","DOI":"10.52202\/079017-2895","article-title":"Learning structured representations with hyperbolic embeddings","volume":"37","author":"Sinha","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109172_bib0034","first-page":"1243","article-title":"Generalization bounds for graph embedding using negative sampling: Linear vs hyperbolic","volume":"34","author":"Suzuki","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109172_bib0035","series-title":"International conference on learning representations (ICLR)","article-title":"Graph attention networks","author":"Velickovic","year":"2017"},{"issue":"1","key":"10.1016\/j.neunet.2026.109172_bib0036","doi-asserted-by":"crossref","first-page":"4069","DOI":"10.1038\/s41467-020-17866-2","article-title":"Brain-inspired replay for continual learning with artificial neural networks","volume":"11","author":"Van de Ven","year":"2020","journal-title":"Nature Communications"},{"key":"10.1016\/j.neunet.2026.109172_bib0037","series-title":"High-dimensional probability: An introduction with applications in data science","volume":"vol. 47","author":"Vershynin","year":"2018"},{"issue":"12","key":"10.1016\/j.neunet.2026.109172_bib0038","doi-asserted-by":"crossref","first-page":"5079","DOI":"10.1109\/TNNLS.2019.2963066","article-title":"Pid controller-based stochastic optimization acceleration for deep neural networks","volume":"31","author":"Wang","year":"2020","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"issue":"1","key":"10.1016\/j.neunet.2026.109172_bib0039","doi-asserted-by":"crossref","first-page":"396","DOI":"10.1162\/qss_a_00021","article-title":"Microsoft academic graph: When experts are not enough","volume":"1","author":"Wang","year":"2020","journal-title":"Quantitative Science Studies"},{"key":"10.1016\/j.neunet.2026.109172_bib0040","first-page":"3338","article-title":"Adversarial robustness in graph neural networks: A hamiltonian approach","volume":"36","author":"Zhao","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"10.1016\/j.neunet.2026.109172_bib0041","series-title":"2022\u202fIEEE International conference on autonomic computing and self-organizing systems (ACSOS)","first-page":"71","article-title":"FedACA: An adaptive communication-efficient asynchronous framework for federated learning","author":"Zhou","year":"2022"}],"container-title":["Neural Networks"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0893608026006337?httpAccept=text\/xml","content-type":"text\/xml","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/api.elsevier.com\/content\/article\/PII:S0893608026006337?httpAccept=text\/plain","content-type":"text\/plain","content-version":"vor","intended-application":"text-mining"}],"deposited":{"date-parts":[[2026,6,4]],"date-time":"2026-06-04T17:58:26Z","timestamp":1780595906000},"score":1,"resource":{"primary":{"URL":"https:\/\/linkinghub.elsevier.com\/retrieve\/pii\/S0893608026006337"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,11]]},"references-count":41,"alternative-id":["S0893608026006337"],"URL":"https:\/\/doi.org\/10.1016\/j.neunet.2026.109172","relation":{},"ISSN":["0893-6080"],"issn-type":[{"value":"0893-6080","type":"print"}],"subject":[],"published":{"date-parts":[[2026,11]]},"assertion":[{"value":"Elsevier","name":"publisher","label":"This article is maintained by"},{"value":"When task performance deceives: Task-geometry decoupling in learnable-curvature hyperbolic GNNs","name":"articletitle","label":"Article Title"},{"value":"Neural Networks","name":"journaltitle","label":"Journal Title"},{"value":"https:\/\/doi.org\/10.1016\/j.neunet.2026.109172","name":"articlelink","label":"CrossRef DOI link to publisher maintained version"},{"value":"article","name":"content_type","label":"Content Type"},{"value":"\u00a9 2026 Elsevier Ltd. All rights are reserved, including those for text and data mining, AI training, and similar technologies.","name":"copyright","label":"Copyright"}],"article-number":"109172"}}