{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T12:15:17Z","timestamp":1776341717751,"version":"3.51.2"},"reference-count":53,"publisher":"Springer Science and Business Media LLC","issue":"2","license":[{"start":{"date-parts":[[2026,2,18]],"date-time":"2026-02-18T00:00:00Z","timestamp":1771372800000},"content-version":"tdm","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"},{"start":{"date-parts":[[2026,3,12]],"date-time":"2026-03-12T00:00:00Z","timestamp":1773273600000},"content-version":"vor","delay-in-days":22,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0"}],"funder":[{"DOI":"10.13039\/100005326","name":"Yale University","doi-asserted-by":"publisher","id":[{"id":"10.13039\/100005326","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":["link.springer.com"],"crossmark-restriction":false},"short-container-title":["Neural Process Lett"],"abstract":"<jats:title>Abstract<\/jats:title>\n                  <jats:p>Graph machine learning has enjoyed a meteoric rise in popularity since the introduction of deep learning in graph contexts. This is no surprise due to the ubiquity of graph data in large scale industrial settings. Tacitly assumed in all graph learning tasks is the separation of the graph structure and node features: node features strictly encode individual data while the graph structure consists only of pairwise interactions. The driving belief is that node features are (by themselves) insufficient for these tasks, so benchmark performance accurately reflects improvements in graph learning. In our paper, we challenge this orthodoxy by showing that, surprisingly, node features are oftentimes more-than-sufficient for many common graph benchmarks, breaking this critical assumption. When comparing against a well-tuned feature-only MLP baseline on seven of the most commonly used graph learning datasets, one gains little benefit from using graph structure on five datasets. We posit that these datasets do not benefit considerably from graph learning because the features themselves already contain enough graph information to obviate or substantially reduce the need for the graph. To illustrate this point, we perform a feature study on these datasets and show how the features are responsible for closing the gap between MLP and graph-method performance. Further, in service of introducing better empirical measures of progress for graph neural networks, we present a challenging parametric family of principled synthetic datasets that necessitate graph information for nontrivial performance. Lastly, we section out a subset of real-world datasets that are not trivially solved by an MLP and hence serve as reasonable benchmarks for graph neural networks.<\/jats:p>","DOI":"10.1007\/s11063-026-11833-6","type":"journal-article","created":{"date-parts":[[2026,2,18]],"date-time":"2026-02-18T03:47:41Z","timestamp":1771386461000},"update-policy":"https:\/\/doi.org\/10.1007\/springer_crossmark_policy","source":"Crossref","is-referenced-by-count":0,"title":["Revisiting Graph Learning Benchmarks: When is the Graph Actually Necessary?"],"prefix":"10.1007","volume":"58","author":[{"given":"Isay","family":"Katsman","sequence":"first","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ethan","family":"Lou","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anna","family":"Gilbert","sequence":"additional","affiliation":[],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"297","published-online":{"date-parts":[[2026,2,18]]},"reference":[{"key":"11833_CR1","unstructured":"Kipf T, Welling M (2017) Semi-supervised classification with graph convolutional networks. ArXiv: abs\/1609.02907"},{"key":"11833_CR2","unstructured":"Velickovic P, Cucurull G, Casanova A, Romero A, Lio\u2019 P, Bengio Y (2018) Graph attention networks. ArXiv: abs\/1710.10903"},{"key":"11833_CR3","unstructured":"Hamilton WL, Ying Z, Leskovec J (2017) Inductive representation learning on large graphs. In: NIPS"},{"key":"11833_CR4","unstructured":"Xu K, Hu W, Leskovec J, Jegelka S (2018) How powerful are graph neural networks? ArXiv: abs\/1810.00826"},{"key":"11833_CR5","doi-asserted-by":"crossref","unstructured":"Liu M, Gao H, Ji S (2020) Towards deeper graph neural networks. Proceedings of the 26th ACM SIGKDD International Conference on Knowledge Discovery & Data Mining","DOI":"10.1145\/3394486.3403076"},{"key":"11833_CR6","unstructured":"Wu F, Zhang T, Souza AH, Fifty C, Yu T, Weinberger KQ (2019) Simplifying graph convolutional networks. ArXiv: abs\/1902.07153"},{"key":"11833_CR7","unstructured":"Bodnar C, Frasca F, Wang YG, Otter N, Mont\u00fafar G, Lio\u2019 P, Bronstein MM (2021) Weisfeiler and lehman go topological: Message passing simplicial networks. ArXiv: abs\/2103.03212"},{"key":"11833_CR8","unstructured":"Ganea O, B\u00e9cigneul G, Hofmann T (2018) Hyperbolic neural networks. In: Advances in Neural Information Processing Systems, 5345\u20135355"},{"key":"11833_CR9","unstructured":"Di Giovanni F, Giusti L, Barbero F, Luise G, Lio P, Bronstein MM (2023) On over-squashing in message passing neural networks: The impact of width, depth, and topology. In: Krause, A., Brunskill, E., Cho, K., Engelhardt, B., Sabato, S., Scarlett, J. (eds.) Proceedings of the 40th International Conference on Machine Learning. Proceedings of Machine Learning Research, 202, 7865\u20137885. PMLR"},{"key":"11833_CR10","doi-asserted-by":"publisher","first-page":"440","DOI":"10.1038\/30918","volume":"393","author":"DJ Watts","year":"1998","unstructured":"Watts DJ, Strogatz SH (1998) Collective dynamics of \u2018small-world\u2019 networks. Nature 393:440\u2013442","journal-title":"Nature"},{"issue":"3","key":"11833_CR11","doi-asserted-by":"publisher","first-page":"714","DOI":"10.1109\/72.572108","volume":"8","author":"A Sperduti","year":"1997","unstructured":"Sperduti A, Starita A (1997) Supervised neural networks for the classification of structures. IEEE Trans Neural Networks 8(3):714\u201335","journal-title":"IEEE Trans Neural Networks"},{"key":"11833_CR12","doi-asserted-by":"publisher","first-page":"4","DOI":"10.1109\/TNNLS.2020.2978386","volume":"32","author":"Z Wu","year":"2019","unstructured":"Wu Z, Pan S, Chen F, Long G, Zhang C, Yu PS (2019) A comprehensive survey on graph neural networks. IEEE Transactions on Neural Networks and Learning Systems 32:4\u201324","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"11833_CR13","unstructured":"Huang Q, He H, Singh A, Lim S-N, Benson AR (2020) Combining label propagation and simple models out-performs graph neural networks. ArXiv: abs\/2010.13993"},{"key":"11833_CR14","doi-asserted-by":"publisher","first-page":"204","DOI":"10.1016\/j.neucom.2008.07.021","volume":"73","author":"E Trentin","year":"2009","unstructured":"Trentin E, Iorio ED (2009) Classification of graphical data made easy. Neurocomputing 73:204\u2013212","journal-title":"Neurocomputing"},{"key":"11833_CR15","doi-asserted-by":"publisher","first-page":"14","DOI":"10.1016\/j.neucom.2018.05.095","volume":"313","author":"E Trentin","year":"2018","unstructured":"Trentin E, Iorio ED (2018) Nonparametric small random networks for graph-structured pattern recognition. Neurocomputing 313:14\u201324","journal-title":"Neurocomputing"},{"key":"11833_CR16","unstructured":"Shchur O, Mumme M, Bojchevski A, G\u00fcnnemann S (2018) Pitfalls of graph neural network evaluation. ArXiv: abs\/1811.05868"},{"key":"11833_CR17","doi-asserted-by":"crossref","unstructured":"Sen P, Namata G, Bilgic M, Getoor L, Gallagher B, Eliassi-Rad T (2008) Collective classification in network data. In: The AI Magazine. https:\/\/api.semanticscholar.org\/CorpusID:62016134","DOI":"10.1609\/aimag.v29i3.2157"},{"key":"11833_CR18","doi-asserted-by":"crossref","unstructured":"Kleinberg JM (2000) The small-world phenomenon: an algorithmic perspective. In: Symposium on the Theory of Computing. https:\/\/api.semanticscholar.org\/CorpusID:221559836","DOI":"10.1145\/335305.335325"},{"key":"11833_CR19","unstructured":"Oono K, Suzuki T (2019) On asymptotic behaviors of graph cnns from dynamical systems perspective. CoRR abs\/1905.10947"},{"key":"11833_CR20","unstructured":"Alon U, Yahav E (2021) On the bottleneck of graph neural networks and its practical implications. In: International Conference on Learning Representations. https:\/\/openreview.net\/forum?id=i80OPhOCVH2"},{"key":"11833_CR21","unstructured":"Rusch TK, Bronstein MM, Mishra S (2023) A Survey on Oversmoothing in Graph Neural Networks"},{"key":"11833_CR22","unstructured":"Nguyen K, Hieu NM, Nguyen VD, Ho N, Osher S, Nguyen TM (2023) Revisiting over-smoothing and over-squashing using ollivier-ricci curvature. In: International Conference on Machine Learning, 25956\u201325979. PMLR"},{"key":"11833_CR23","unstructured":"Sonthalia R, Gilbert A, Durham M (2023) Relwire: Metric based graph rewiring. In: NeurIPS 2023 Workshop on Symmetry and Geometry in Neural Representations. https:\/\/openreview.net\/forum?id=c9u8tH1WA0"},{"key":"11833_CR24","unstructured":"Gasteiger J, Wei\u00dfenberger S, G\u00fcnnemann S (2019) Diffusion improves graph learning. Advances in neural information processing systems 32"},{"key":"11833_CR25","unstructured":"Karhadkar K, Banerjee PK, Montufar G (2023) FoSR: First-order spectral rewiring for addressing oversquashing in GNNs. In: The Eleventh International Conference on Learning Representations. https:\/\/openreview.net\/forum?id=3YjQfCLdrzz"},{"key":"11833_CR26","unstructured":"Black M, Wan Z, Nayyeri A, Wang Y (2023) Understanding oversquashing in gnns through the lens of effective resistance. In: International Conference on Machine Learning, 2528\u20132547. PMLR"},{"key":"11833_CR27","unstructured":"Arnaiz-Rodriguez A, Begga A, Escolano F, Oliver N (2022) Diffwire: Inductive graph rewiring via the lovasz bound. In: The First Learning on Graphs Conference. https:\/\/openreview.net\/pdf?id=IXvfIex0mX6f"},{"key":"11833_CR28","doi-asserted-by":"publisher","first-page":"18","DOI":"10.1109\/MSP.2017.2693418","volume":"34","author":"M Bronstein","year":"2017","unstructured":"Bronstein M, Bruna J, LeCun Y, Szlam AD, Vandergheynst P (2017) Geometric deep learning: Going beyond euclidean data. IEEE Signal Process Mag 34:18\u201342","journal-title":"IEEE Signal Process Mag"},{"key":"11833_CR29","unstructured":"Nickel M, Kiela D (2017) Poincar\u00e9 embeddings for learning hierarchical representations. In: Advances in Neural Information Processing Systems, 6338\u20136347"},{"key":"11833_CR30","unstructured":"Chami I, Ying Z, R\u00e9 C, Leskovec J (2019) Hyperbolic graph convolutional neural networks. In: Advances in Neural Information Processing Systems, 4868\u20134879"},{"key":"11833_CR31","unstructured":"Lou A, Katsman I, Jiang Q, Belongie S, Lim S-N, De Sa C (2020) Differentiating through the fr\u00e9chet mean. In: International Conference on Machine Learning"},{"key":"11833_CR32","unstructured":"Chen W, Han X, Lin Y, Zhao H, Liu Z, Li P, Sun M, Zhou J (2021) Fully hyperbolic neural networks. ArXiv: abs\/2105.14686"},{"key":"11833_CR33","first-page":"1690","volume":"8","author":"Y Zhang","year":"2019","unstructured":"Zhang Y, Wang X, Jiang X, Shi C, Ye Y (2019) Hyperbolic graph attention network. IEEE Transactions on Big Data 8:1690\u20131701","journal-title":"IEEE Transactions on Big Data"},{"key":"11833_CR34","doi-asserted-by":"crossref","unstructured":"Zhang Y, Wang X, Shi C, Liu N, Song G (2021) Lorentzian graph convolutional networks. Proceedings of the Web Conference 2021","DOI":"10.1145\/3442381.3449872"},{"key":"11833_CR35","doi-asserted-by":"crossref","unstructured":"Sarkar R (2011) Low distortion delaunay embedding of trees in hyperbolic plane. In: International Symposium on Graph Drawing, 355\u2013366. Springer","DOI":"10.1007\/978-3-642-25878-7_34"},{"key":"11833_CR36","unstructured":"Luo, Y., Chen, A., Yan, K., Tian, L.: Distilling self-knowledge from contrastive links to classify graph nodes without passing messages. ArXiv: abs\/2106.08541 (2021)"},{"key":"11833_CR37","unstructured":"Zhang S, Liu Y, Sun Y, Shah N (2021) Graph-less neural networks: Teaching old mlps new tricks via distillation. ArXiv: abs\/2110.08727"},{"key":"11833_CR38","unstructured":"Tortorella D, Micheli A (2023) Is Rewiring Actually Helpful in Graph Neural Networks?"},{"key":"11833_CR39","unstructured":"Bechler-Speicher M, Amos I, Gilad-Bachrach R, Globerson A (2024) Graph Neural Networks Use Graphs When They Shouldn\u2019t"},{"key":"11833_CR40","unstructured":"Katsman I, Gilbert A (2024) Shedding light on problems with hyperbolic graph learning. https:\/\/api.semanticscholar.org\/CorpusID:273963557"},{"key":"11833_CR41","doi-asserted-by":"crossref","unstructured":"Thomas P (2009) Semi-supervised learning (chapelle, o. et al., eds.; 2006) [book reviews]. IEEE Transactions on Neural Networks 20, 542\u2013542","DOI":"10.1109\/TNN.2009.2015974"},{"key":"11833_CR42","doi-asserted-by":"crossref","unstructured":"Monti F, Boscaini D, Masci J, Rodol\u00e0 E, Svoboda J, Bronstein MM (2016) Geometric deep learning on graphs and manifolds using mixture model cnns. 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 5425\u20135434","DOI":"10.1109\/CVPR.2017.576"},{"key":"11833_CR43","unstructured":"Hoang VT, Lee O-J (2023) Mitigating degree biases in message passing mechanism by utilizing community structures. ArXiv: abs\/2312.16788"},{"key":"11833_CR44","unstructured":"Shirzad H, Velingker A, Venkatachalam B, Sutherland DJ, Sinop AK (2023) Exphormer: Sparse transformers for graphs. In: International Conference on Machine Learning. https:\/\/api.semanticscholar.org\/CorpusID:257482539"},{"key":"11833_CR45","unstructured":"Wang M, Zheng D, Ye Z, Gan Q, Li M, Song X, Zhou J, Ma C, Yu L, Gai Y, Xiao T, He T, Karypis G, Li J, Zhang Z (2019) Deep graph library: A graph-centric, highly-performant package for graph neural networks. arXiv preprint arXiv:1909.01315"},{"key":"11833_CR46","unstructured":"Defferrard M, Bresson X, Vandergheynst P (2016) Convolutional neural networks on graphs with fast localized spectral filtering. In: Neural Information Processing Systems. https:\/\/api.semanticscholar.org\/CorpusID:3016223"},{"key":"11833_CR47","unstructured":"Klicpera J, Bojchevski A, G\u00fcnnemann S (2018) Predict then propagate: Graph neural networks meet personalized pagerank. In: International Conference on Learning Representations. https:\/\/api.semanticscholar.org\/CorpusID:67855539"},{"key":"11833_CR48","doi-asserted-by":"crossref","unstructured":"Izadi MR, Fang Y, Stevenson RL, Lin L (2020) Optimization of graph neural networks with natural gradient descent. 2020 IEEE International Conference on Big Data (Big Data), 171\u2013179","DOI":"10.1109\/BigData50022.2020.9378063"},{"key":"11833_CR49","unstructured":"Luan S, Hua C, Lu Q, Zhu J, Zhao M, Zhang S, Chang X, Precup D (2021) Is heterophily a real nightmare for graph neural networks to do node classification? ArXiv: abs\/2109.05641"},{"key":"11833_CR50","doi-asserted-by":"publisher","first-page":"5441","DOI":"10.1109\/TNNLS.2018.2803523","volume":"29","author":"M Bongini","year":"2018","unstructured":"Bongini M, Rigutini L, Trentin E (2018) Recursive neural networks for density estimation over generalized random graphs. IEEE Transactions on Neural Networks and Learning Systems 29:5441\u20135458","journal-title":"IEEE Transactions on Neural Networks and Learning Systems"},{"key":"11833_CR51","unstructured":"Rozemberczki B, Allen C, Sarkar R (2019) Multi-scale attributed node embedding. ArXiv: abs\/1909.13021"},{"key":"11833_CR52","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: abs\/2005.00687"},{"key":"11833_CR53","doi-asserted-by":"crossref","unstructured":"Harris CR, Millman KJ, Walt S, Gommers R, Virtanen P, Cournapeau D, Wieser E, Taylor J, Berg S, Smith NJ, Kern R, Picus M, Hoyer S, Kerkwijk MH, Brett M, Haldane A, R\u2019io JF, Wiebe M, Peterson P, G\u2019erard-Marchant P, Sheppard K, Reddy T, Weckesser W, Abbasi H, Gohlke C, Oliphant TE (2020) Array programming with numpy Nature 585:357\u2013362","DOI":"10.1038\/s41586-020-2649-2"}],"container-title":["Neural Processing Letters"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/link.springer.com\/article\/10.1007\/s11063-026-11833-6","content-type":"text\/html","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11063-026-11833-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/link.springer.com\/content\/pdf\/10.1007\/s11063-026-11833-6.pdf","content-type":"application\/pdf","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,4,16]],"date-time":"2026-04-16T11:35:43Z","timestamp":1776339343000},"score":1,"resource":{"primary":{"URL":"https:\/\/link.springer.com\/10.1007\/s11063-026-11833-6"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,2,18]]},"references-count":53,"journal-issue":{"issue":"2","published-online":{"date-parts":[[2026,4]]}},"alternative-id":["11833"],"URL":"https:\/\/doi.org\/10.1007\/s11063-026-11833-6","relation":{},"ISSN":["1573-773X"],"issn-type":[{"value":"1573-773X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,2,18]]},"assertion":[{"value":"4 August 2025","order":1,"name":"received","label":"Received","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"4 January 2026","order":2,"name":"accepted","label":"Accepted","group":{"name":"ArticleHistory","label":"Article History"}},{"value":"18 February 2026","order":3,"name":"first_online","label":"First Online","group":{"name":"ArticleHistory","label":"Article History"}},{"order":1,"name":"Ethics","group":{"name":"EthicsHeading","label":"Declarations"}},{"value":"The authors declare no competing interests.","order":2,"name":"Ethics","group":{"name":"EthicsHeading","label":"Competing interests"}}],"article-number":"24"}}