{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T08:08:26Z","timestamp":1774339706168,"version":"3.50.1"},"reference-count":27,"publisher":"MDPI AG","issue":"3","license":[{"start":{"date-parts":[[2026,3,21]],"date-time":"2026-03-21T00:00:00Z","timestamp":1774051200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/"}],"funder":[{"DOI":"10.13039\/501100012216","name":"Jiangsu University of Science and Technology","doi-asserted-by":"crossref","award":["1022932208"],"award-info":[{"award-number":["1022932208"]}],"id":[{"id":"10.13039\/501100012216","id-type":"DOI","asserted-by":"crossref"}]},{"DOI":"10.13039\/501100012216","name":"Jiangsu University of Science and Technology","doi-asserted-by":"crossref","award":["1132932301"],"award-info":[{"award-number":["1132932301"]}],"id":[{"id":"10.13039\/501100012216","id-type":"DOI","asserted-by":"crossref"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["Entropy"],"abstract":"<jats:p>Graph Neural Networks (GNNs) have emerged as the standard for learning representations from graph-structured data. While traditional architectures relying on message-passing mechanisms excel in homophilic settings, they essentially function as fixed low-pass filters. However, this smoothing operation limits their ability to generalize to heterophilic graphs, where connected nodes often exhibit dissimilar labels and high-frequency signals are crucial for discrimination. Furthermore, existing Mixture-of-Experts (MoE) methods for graphs often suffer from local-view routing, failing to capture global structural context during expert selection. To address these challenges, this paper proposes SS-AdaMoE, a novel Spatio-Spectral Adaptive Mixture of Experts framework designed for robust node classification across diverse graph patterns. Specifically, a Dual-Domain Expert System is constructed, integrating heterogeneous spatial aggregators with learnable spectral filters based on Bernstein polynomials. This allows the model to adaptively capture arbitrary frequency responses\u2014including high-pass and band-pass signals\u2014which are overlooked by standard GNNs. To resolve the locality bias, a Hierarchical Global-Prior Gating Network augmented by a Linear Graph Transformer is introduced, ensuring that expert selection is guided by both local node features and global topological awareness. Extensive experiments are conducted on five benchmark datasets spanning both homophilic and heterophilic networks. The results demonstrate that SS-AdaMoE consistently outperforms baselines, achieving accuracy improvements of up to 2.65% on Chameleon and 1.41% on Roman-empire over the strongest MoE baseline, while surpassing traditional GCN architectures by margins exceeding 28% on heterophilic datasets such as Texas. These findings validate that the synergy of learnable spectral priors and global gating effectively bridges the gap between spatial aggregation and spectral filtering.<\/jats:p>","DOI":"10.3390\/e28030355","type":"journal-article","created":{"date-parts":[[2026,3,23]],"date-time":"2026-03-23T11:59:36Z","timestamp":1774267176000},"page":"355","update-policy":"https:\/\/doi.org\/10.3390\/mdpi_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["SS-AdaMoE: Spatio-Spectral Adaptive Mixture of Experts with Global Structural Priors for Graph Node Classification"],"prefix":"10.3390","volume":"28","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-4469-0578","authenticated-orcid":false,"given":"Xilin","family":"Kang","sequence":"first","affiliation":[{"name":"School of Computer, Jiangsu University of Science and Technology, Zhenjiang 212100, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tianyue","family":"Yu","sequence":"additional","affiliation":[{"name":"School of Computer, Jiangsu University of Science and Technology, Zhenjiang 212100, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Letao","family":"Wang","sequence":"additional","affiliation":[{"name":"School of Foreign Languages, Jiangsu University, Zhenjiang 212013, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yutong","family":"Guo","sequence":"additional","affiliation":[{"name":"School of Computer, Jiangsu University of Science and Technology, Zhenjiang 212100, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5850-1922","authenticated-orcid":false,"given":"Fengjun","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer, Jiangsu University of Science and Technology, Zhenjiang 212100, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"1968","published-online":{"date-parts":[[2026,3,21]]},"reference":[{"key":"ref_1","unstructured":"Kipf, T.N., and Welling, M. (2017, January 24\u201326). Semi-Supervised Classification with Graph Convolutional Networks. Proceedings of the International Conference on Learning Representations (ICLR), Toulon, France."},{"key":"ref_2","unstructured":"Veli\u010dkovi\u0107, P., Cucurull, G., Casanova, A., Romero, A., Lio, P., and Bengio, Y. (May, January 30). Graph Attention Networks. Proceedings of the International Conference on Learning Representations (ICLR), Vancouver, BC, Canada."},{"key":"ref_3","unstructured":"Zhu, J., Yan, Y., Zhao, L., Heimann, M., Akoglu, L., and Koutra, D. (2020, January 6\u201312). Beyond homophily in graph neural networks: Current limitations and effective designs. Proceedings of the 34th International Conference on Neural Information Processing Systems, Vancouver, BC, Canada."},{"key":"ref_4","unstructured":"Yin, H., Zhang, M., Wang, Y., Li, J., and Pan, S. (2023, January 1\u20135). Graph Mixture of Experts: Learning on Large-Scale Graphs with Explicit Diversity Modeling. Proceedings of the International Conference on Learning Representations (ICLR), Kigali, Rwanda."},{"key":"ref_5","unstructured":"Han, H., Liu, X., Feng, F., and Chua, T.S. (2024, January 21\u201327). Node-wise Filtering in Graph Neural Networks: A Mixture of Experts Approach. Proceedings of the International Conference on Machine Learning (ICML), Vienna, Austria."},{"key":"ref_6","unstructured":"Zhang, J., and Smith, J. (2024). Mixture of Experts for Node Classification. arXiv."},{"key":"ref_7","unstructured":"Wang, X., and Zhang, M. (2022, January 17\u201323). How Powerful are Spectral Graph Neural Networks. Proceedings of the International Conference on Machine Learning (ICML), Baltimore, MD, USA."},{"key":"ref_8","unstructured":"Wu, Q., Zhao, W., Li, Z., Wipf, D., and Yan, J. (December, January 28). NodeFormer: A Scalable Graph Structure Learning Transformer for Node Classification. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), New Orleans, LA, USA."},{"key":"ref_9","unstructured":"Pei, H., Wei, B., Chang, K.C.C., Lei, Y., and Yang, B. (2020, January 26\u201330). Geom-GCN: Geometric Graph Convolutional Networks. Proceedings of the 8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia."},{"key":"ref_10","unstructured":"Defferrard, M., Bresson, X., and Vandergheynst, P. (2016, January 5\u201310). Convolutional Neural Networks on Graphs with Fast Localized Spectral Filtering. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Barcelona, Spain."},{"key":"ref_11","unstructured":"Wang, T., Jin, D., Wang, R., He, D., and Huang, Y. (March, January 22). Powerful graph convolutional networks with adaptive propagation mechanism for homophily and heterophily. Proceedings of the AAAI Conference on Artificial Intelligence, Online."},{"key":"ref_12","unstructured":"Chien, E., Peng, J., Li, P., and Milenkovic, O. (2021, January 3\u20137). Adaptive Universal Generalized PageRank Graph Neural Network. Proceedings of the International Conference on Learning Representations (ICLR), Virtual Event."},{"key":"ref_13","unstructured":"Bo, D., Wang, X., Shi, C., and Shen, H. (2021, January 2\u20139). Beyond low-frequency information in graph convolutional networks. Proceedings of the AAAI Conference on Artificial Intelligence, Virtual."},{"key":"ref_14","first-page":"1362","article-title":"Revisiting heterophily for graph neural networks","volume":"35","author":"Luan","year":"2022","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref_15","unstructured":"Chen, D., Pellizzoni, P., and Borgwardt, K. (2024, January 10\u201315). Spatio-Spectral Graph Neural Networks. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Vancouver, BC, Canada."},{"key":"ref_16","unstructured":"Luan, S., Hua, C., Lu, Q., Ma, L., Wu, L., Wang, X., Xu, M., Chang, X.-W., Precup, D., and Ying, R. (2024). The Heterophilic Graph Learning Handbook: Benchmarks, Models, Theoretical Analysis, Applications and Challenges. arXiv."},{"key":"ref_17","unstructured":"Ramp\u00e1\u0161ek, L., Galkin, M., Dwivedi, V.P., Luu, A.T., Wolf, G., and Beressy, D. (December, January 28). Recipe for a General, Powerful, Scalable Graph Transformer. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), New Orleans, LA, USA."},{"key":"ref_18","unstructured":"Bo, D., Shi, C., Wang, L., and Liao, R. (2023, January 5). Specformer: Spectral Graph Neural Networks Meet Transformers. Proceedings of the ICLR; OpenReview.net, Kigali, Rwanda."},{"key":"ref_19","unstructured":"Li, P., Zhang, Z., Yadav, P., Sung, Y.L., Cheng, Y., Bansal, M., and Chen, T. (2024, January 7\u201311). Merge, Then Compress: Demystify Efficient SMoE with Hints from Its Routing Policy. Proceedings of the International Conference on Learning Representations (ICLR), Vienna, Austria."},{"key":"ref_20","unstructured":"Hamilton, W.L., Ying, R., and Leskovec, J. (2017, January 4\u20139). Inductive representation learning on large graphs. Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, CA, USA. NIPS\u201917."},{"key":"ref_21","first-page":"5453","article-title":"Representation Learning on Graphs with Jumping Knowledge Networks","volume":"Volume 80","author":"Dy","year":"2018","journal-title":"Proceedings of the 35th International Conference on Machine Learning, Stockholm, Sweden, 10\u201315 July 2018"},{"key":"ref_22","first-page":"5156","article-title":"Transformers are RNNs: Fast Autoregressive Transformers with Linear Attention","volume":"Volume 119","author":"Katharopoulos","year":"2020","journal-title":"Proceedings of the 37th International Conference on Machine Learning, Virtual, 13\u201318 July 2020"},{"key":"ref_23","doi-asserted-by":"crossref","first-page":"127","DOI":"10.1023\/A:1009953814988","article-title":"Automating the Construction of Internet Portals with Machine Learning","volume":"3","author":"McCallum","year":"2000","journal-title":"Inf. Retr."},{"key":"ref_24","unstructured":"Ghani, R., Jones, R., Mladenic, D., Nigam, K., and Slattery, S. (2000, January 20\u201323). Data mining on symbolic knowledge extracted from the web. Proceedings of the Workshop on Text Mining at the Sixth ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Boston, MA, USA."},{"key":"ref_25","unstructured":"Platonov, O., Kuznedelev, D., Babenko, A., and Prokhorenkova, L. (2023). A Critical Look at the Evaluation of GNNs under Heterophily: Are We Really Making Progress?. arXiv."},{"key":"ref_26","unstructured":"Lim, D., Hohne, F., Li, X., Huang, S.L., Gupta, V., Bhalerao, O., and Lim, S.N. (2021, January 6\u201314). Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methods. Proceedings of the Advances in Neural Information Processing Systems (NeurIPS), Online."},{"key":"ref_27","unstructured":"Li, X., Zhu, R., Cheng, Y., Shan, C., Luo, S., Li, D., and Qian, W. (2022, January 25\u201329). GloGNN: Global Structural Information for Effective Node Classification in Heterophilic Graphs. Proceedings of the ACM Web Conference, Virtual Event, Lyon, France."}],"container-title":["Entropy"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mdpi.com\/1099-4300\/28\/3\/355\/pdf","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,24]],"date-time":"2026-03-24T05:11:59Z","timestamp":1774329119000},"score":1,"resource":{"primary":{"URL":"https:\/\/www.mdpi.com\/1099-4300\/28\/3\/355"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,21]]},"references-count":27,"journal-issue":{"issue":"3","published-online":{"date-parts":[[2026,3]]}},"alternative-id":["e28030355"],"URL":"https:\/\/doi.org\/10.3390\/e28030355","relation":{},"ISSN":["1099-4300"],"issn-type":[{"value":"1099-4300","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,3,21]]}}}