{"url":"/sota/graph-classification-on-ptc","task":{"name":"Graph Classification","url":"/task/graph-classification","note":null},"dataset":{"name":"PTC","url":"/dataset/ptc"},"category":"Graphs","categories":["Graphs"],"category_note":null,"description":"**Graph Classification** is a task that involves classifying a graph-structured data into different classes or categories. Graphs are a powerful way to represent relationships and interactions between different entities, and graph classification can be applied to a wide range of applications, such as social network analysis, bioinformatics, and recommendation systems. In graph classification, the input is a graph, and the goal is to learn a classifier that can accurately predict the class of the graph.\r\n\r\n<span style=\"color:grey; opacity: 0.6\">( Image credit: [Hierarchical Graph Pooling with Structure Learning](https://github.com/cszhangzhen/HGP-SL) )</span>","description_from":"task","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","rank":"the archive's row order at snapshot; not re-ranked","rows_end_at":"2025-07-28","rows_withheld_as_spam":0,"metric_values":"the archive's strings, untouched"},"metrics":["Accuracy"],"metric_direction":{"note":"inferred from the metric name only (the archive records no direction); null = not inferred, chart draws points only","by_metric":{"Accuracy":"higher"}},"counts":{"rows":37,"rows_with_code":30,"rows_with_paper_page":37,"rows_dated":37,"rows_using_additional_data":1},"rows":[{"rank_in_archive_order":1,"model":"U2GNN (Unsupervised)","metrics":{"Accuracy":"91.81%"},"uses_additional_data":false,"paper_date":"2019-09-26","paper":"/paper/unsupervised-universal-self-attention-network","paper_url":"https://arxiv.org/abs/1909.11855v9","paper_title":"Universal Graph Transformer Self-Attention Networks","code":"https://github.com/daiquocnguyen/Graph-Transformer","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":0,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":2,"model":"GIUNet","metrics":{"Accuracy":"85.7%"},"uses_additional_data":false,"paper_date":"2023-08-23","paper":"/paper/graph-isomorphism-unet","paper_url":"https://www.sciencedirect.com/science/article/abs/pii/S0957417423017827#preview-section-abstract","paper_title":"Graph isomorphism UNet","code":"https://github.com/aramiracle/GIUNet","n_code_links":1,"syntology":null},{"rank_in_archive_order":3,"model":"GIC","metrics":{"Accuracy":"77.64%"},"uses_additional_data":false,"paper_date":"2018-11-11","paper":"/paper/gaussian-induced-convolution-for-graphs","paper_url":"http://arxiv.org/abs/1811.04393v1","paper_title":"Gaussian-Induced Convolution for Graphs","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":4,"model":"DUGNN","metrics":{"Accuracy":"74.7%"},"uses_additional_data":true,"paper_date":"2019-09-22","paper":"/paper/190910086","paper_url":"https://arxiv.org/abs/1909.10086v3","paper_title":"Learning Universal Graph Neural Network Embeddings With Aid Of Transfer Learning","code":"https://github.com/vermaMachineLearning/Universal-Graph-Embedding-Neural-Network","n_code_links":1,"syntology":null},{"rank_in_archive_order":5,"model":"UGraphEmb-F","metrics":{"Accuracy":"73.56%"},"uses_additional_data":false,"paper_date":"2019-04-01","paper":"/paper/unsupervised-inductive-whole-graph-embedding","paper_url":"https://arxiv.org/abs/1904.01098v2","paper_title":"Unsupervised Inductive Graph-Level Representation Learning via Graph-Graph Proximity","code":"https://github.com/yunshengb/UGraphEmb","n_code_links":1,"syntology":null},{"rank_in_archive_order":6,"model":"sGIN","metrics":{"Accuracy":"73.56%"},"uses_additional_data":false,"paper_date":"2019-05-21","paper":"/paper/neighborhood-enlargement-in-graph-neural","paper_url":"https://arxiv.org/abs/1905.08509v4","paper_title":"Mutual Information Maximization in Graph Neural Networks","code":"https://github.com/CODE-SUBMIT/Graph_Neighborhood_1","n_code_links":2,"syntology":null},{"rank_in_archive_order":7,"model":"G_DenseNet","metrics":{"Accuracy":"73.24%"},"uses_additional_data":false,"paper_date":"2018-07-07","paper":"/paper/when-work-matters-transforming-classical","paper_url":"http://arxiv.org/abs/1807.02653v1","paper_title":"When Work Matters: Transforming Classical Network Structures to Graph CNN","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":8,"model":"CIN++","metrics":{"Accuracy":"73.2%"},"uses_additional_data":false,"paper_date":"2023-06-06","paper":"/paper/cin-enhancing-topological-message-passing","paper_url":"https://arxiv.org/abs/2306.03561v1","paper_title":"CIN++: Enhancing Topological Message Passing","code":"https://github.com/twitter-research/cwn","n_code_links":1,"syntology":null},{"rank_in_archive_order":9,"model":"CAN","metrics":{"Accuracy":"72.8%"},"uses_additional_data":false,"paper_date":"2022-09-16","paper":"/paper/cell-attention-networks","paper_url":"https://arxiv.org/abs/2209.08179v1","paper_title":"Cell Attention Networks","code":"https://github.com/lrnzgiusti/can","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":3,"n_samples":5,"n_pointer_only_licence":0}},{"rank_in_archive_order":10,"model":"UGraphEmb","metrics":{"Accuracy":"72.54%"},"uses_additional_data":false,"paper_date":"2019-04-01","paper":"/paper/unsupervised-inductive-whole-graph-embedding","paper_url":"https://arxiv.org/abs/1904.01098v2","paper_title":"Unsupervised Inductive Graph-Level Representation Learning via Graph-Graph Proximity","code":"https://github.com/yunshengb/UGraphEmb","n_code_links":1,"syntology":null},{"rank_in_archive_order":11,"model":"TFGW ADJ (L=2)","metrics":{"Accuracy":"72.4%"},"uses_additional_data":false,"paper_date":"2022-05-31","paper":"/paper/template-based-graph-neural-network-with","paper_url":"https://arxiv.org/abs/2205.15733v1","paper_title":"Template based Graph Neural Network with Optimal Transport Distances","code":"https://github.com/cedricvincentcuaz/TFGW","n_code_links":1,"syntology":null},{"rank_in_archive_order":12,"model":"DGA","metrics":{"Accuracy":"71.24%"},"uses_additional_data":false,"paper_date":"2018-11-17","paper":"/paper/discriminative-graph-autoencoder","paper_url":"https://doi.org/10.1109/ICBK.2018.00033","paper_title":"Discriminative Graph Autoencoder","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":13,"model":"U2GNN","metrics":{"Accuracy":"69.63%"},"uses_additional_data":false,"paper_date":"2019-09-26","paper":"/paper/unsupervised-universal-self-attention-network","paper_url":"https://arxiv.org/abs/1909.11855v9","paper_title":"Universal Graph Transformer Self-Attention Networks","code":"https://github.com/daiquocnguyen/Graph-Transformer","n_code_links":1,"syntology":{"n_ran":4,"n_unverified":0,"n_samples":4,"n_pointer_only_licence":0}},{"rank_in_archive_order":14,"model":"BC + Capsules","metrics":{"Accuracy":"69%"},"uses_additional_data":false,"paper_date":"2019-02-22","paper":"/paper/capsule-neural-networks-for-graph","paper_url":"http://arxiv.org/abs/1902.08399v1","paper_title":"Capsule Neural Networks for Graph Classification using Explicit Tensorial Graph Representations","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":15,"model":"SEG-BERT","metrics":{"Accuracy":"68.86%"},"uses_additional_data":false,"paper_date":"2020-02-09","paper":"/paper/segmented-graph-bert-for-graph-instance","paper_url":"https://arxiv.org/abs/2002.03283v1","paper_title":"Segmented Graph-Bert for Graph Instance Modeling","code":"https://github.com/jwzhanggy/SEG-BERT","n_code_links":1,"syntology":null},{"rank_in_archive_order":16,"model":"Spec-GN","metrics":{"Accuracy":"68.05%"},"uses_additional_data":false,"paper_date":"2021-12-14","paper":"/paper/improving-spectral-graph-convolution-for","paper_url":"https://arxiv.org/abs/2112.07160v2","paper_title":"A New Perspective on the Effects of Spectrum in Graph Neural Networks","code":"https://github.com/qslim/gnn-spectrum","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":2,"n_samples":2,"n_pointer_only_licence":0}},{"rank_in_archive_order":17,"model":"WEGL","metrics":{"Accuracy":"67.5%"},"uses_additional_data":false,"paper_date":"2020-06-16","paper":"/paper/wasserstein-embedding-for-graph-learning","paper_url":"https://arxiv.org/abs/2006.09430v2","paper_title":"Wasserstein Embedding for Graph Learning","code":"https://github.com/navid-naderi/WEGL","n_code_links":1,"syntology":null},{"rank_in_archive_order":18,"model":"WWL","metrics":{"Accuracy":"66.31%"},"uses_additional_data":false,"paper_date":"2019-06-04","paper":"/paper/wasserstein-weisfeiler-lehman-graph-kernels","paper_url":"https://arxiv.org/abs/1906.01277v2","paper_title":"Wasserstein Weisfeiler-Lehman Graph Kernels","code":"https://github.com/BorgwardtLab/WWL","n_code_links":2,"syntology":{"n_ran":1,"n_unverified":10,"n_samples":11,"n_pointer_only_licence":0}},{"rank_in_archive_order":19,"model":"DropGIN","metrics":{"Accuracy":"66.3%"},"uses_additional_data":false,"paper_date":"2021-11-11","paper":"/paper/dropgnn-random-dropouts-increase-the","paper_url":"https://arxiv.org/abs/2111.06283v1","paper_title":"DropGNN: Random Dropouts Increase the Expressiveness of Graph Neural Networks","code":"https://github.com/karolismart/dropgnn","n_code_links":1,"syntology":{"n_ran":2,"n_unverified":1,"n_samples":3,"n_pointer_only_licence":0}},{"rank_in_archive_order":20,"model":"PPGN","metrics":{"Accuracy":"66.17%"},"uses_additional_data":false,"paper_date":"2019-05-27","paper":"/paper/provably-powerful-graph-networks","paper_url":"https://arxiv.org/abs/1905.11136v4","paper_title":"Provably Powerful Graph Networks","code":"https://github.com/hadarser/ProvablyPowerfulGraphNetworks_torch","n_code_links":2,"syntology":{"n_ran":2,"n_unverified":6,"n_samples":8,"n_pointer_only_licence":0}},{"rank_in_archive_order":21,"model":"Deep WL SGN(0,1,2)","metrics":{"Accuracy":"65.88%"},"uses_additional_data":false,"paper_date":"2019-03-21","paper":"/paper/subgraph-networks-with-application-to","paper_url":"https://arxiv.org/abs/1903.09022v3","paper_title":"Subgraph Networks with Application to Structural Feature Space Expansion","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":22,"model":"DGCNN","metrics":{"Accuracy":"65.43%"},"uses_additional_data":false,"paper_date":"2017-12-10","paper":"/paper/dgcnn-disordered-graph-convolutional-neural","paper_url":"http://arxiv.org/abs/1712.03563v1","paper_title":"DGCNN: Disordered Graph Convolutional Neural Network Based on the Gaussian Mixture Model","code":null,"n_code_links":0,"syntology":null},{"rank_in_archive_order":23,"model":"hGANet","metrics":{"Accuracy":"65.02%"},"uses_additional_data":false,"paper_date":"2019-07-05","paper":"/paper/graph-representation-learning-via-hard-and","paper_url":"https://arxiv.org/abs/1907.04652v1","paper_title":"Graph Representation Learning via Hard and Channel-Wise Attention Networks","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/hardgat","n_code_links":1,"syntology":null},{"rank_in_archive_order":24,"model":"GIN-0","metrics":{"Accuracy":"64.40%"},"uses_additional_data":false,"paper_date":"2018-10-01","paper":"/paper/how-powerful-are-graph-neural-networks","paper_url":"http://arxiv.org/abs/1810.00826v3","paper_title":"How Powerful are Graph Neural Networks?","code":"https://github.com/dmlc/dgl/tree/master/examples/mxnet/gin","n_code_links":19,"syntology":{"n_ran":3,"n_unverified":7,"n_samples":10,"n_pointer_only_licence":5}},{"rank_in_archive_order":25,"model":"cGANet","metrics":{"Accuracy":"63.53%"},"uses_additional_data":false,"paper_date":"2019-07-05","paper":"/paper/graph-representation-learning-via-hard-and","paper_url":"https://arxiv.org/abs/1907.04652v1","paper_title":"Graph Representation Learning via Hard and Channel-Wise Attention Networks","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/hardgat","n_code_links":1,"syntology":null},{"rank_in_archive_order":26,"model":"DDGK","metrics":{"Accuracy":"63.14%"},"uses_additional_data":false,"paper_date":"2019-04-21","paper":"/paper/ddgk-learning-graph-representations-for-deep","paper_url":"http://arxiv.org/abs/1904.09671v1","paper_title":"DDGK: Learning Graph Representations for Deep Divergence Graph Kernels","code":"https://github.com/google-research/google-research/tree/master/graph_embedding/ddgk","n_code_links":1,"syntology":null},{"rank_in_archive_order":27,"model":"GANet","metrics":{"Accuracy":"62.94%"},"uses_additional_data":false,"paper_date":"2019-07-05","paper":"/paper/graph-representation-learning-via-hard-and","paper_url":"https://arxiv.org/abs/1907.04652v1","paper_title":"Graph Representation Learning via Hard and Channel-Wise Attention Networks","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/hardgat","n_code_links":1,"syntology":null},{"rank_in_archive_order":28,"model":"DAGCN","metrics":{"Accuracy":"62.88%"},"uses_additional_data":false,"paper_date":"2019-04-04","paper":"/paper/dagcn-dual-attention-graph-convolutional","paper_url":"http://arxiv.org/abs/1904.02278v1","paper_title":"DAGCN: Dual Attention Graph Convolutional Networks","code":"https://github.com/dawenzi123/DAGCN","n_code_links":1,"syntology":null},{"rank_in_archive_order":29,"model":"SF + RFC","metrics":{"Accuracy":"62.8%"},"uses_additional_data":false,"paper_date":"2018-10-22","paper":"/paper/a-simple-baseline-algorithm-for-graph","paper_url":"http://arxiv.org/abs/1810.09155v2","paper_title":"A Simple Baseline Algorithm for Graph Classification","code":"https://github.com/benedekrozemberczki/karateclub","n_code_links":3,"syntology":null},{"rank_in_archive_order":30,"model":"δ-2-LWL","metrics":{"Accuracy":"62.70%"},"uses_additional_data":false,"paper_date":"2019-04-02","paper":"/paper/towards-a-practical-k-dimensional-weisfeiler","paper_url":"https://arxiv.org/abs/1904.01543v3","paper_title":"Weisfeiler and Leman go sparse: Towards scalable higher-order graph embeddings","code":"https://github.com/chrsmrrs/sparsewl","n_code_links":1,"syntology":{"n_ran":0,"n_unverified":1,"n_samples":1,"n_pointer_only_licence":1}},{"rank_in_archive_order":31,"model":"LDP","metrics":{"Accuracy":"61.7%"},"uses_additional_data":false,"paper_date":"2018-11-08","paper":"/paper/a-simple-yet-effective-baseline-for-non","paper_url":"https://arxiv.org/abs/1811.03508v3","paper_title":"A simple yet effective baseline for non-attributed graph classification","code":"https://github.com/benedekrozemberczki/karateclub","n_code_links":2,"syntology":null},{"rank_in_archive_order":32,"model":"InfoGraph","metrics":{"Accuracy":"61.65"},"uses_additional_data":false,"paper_date":"2019-07-31","paper":"/paper/infograph-unsupervised-and-semi-supervised","paper_url":"https://arxiv.org/abs/1908.01000v3","paper_title":"InfoGraph: Unsupervised and Semi-supervised Graph-Level Representation Learning via Mutual Information Maximization","code":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/infograph","n_code_links":5,"syntology":null},{"rank_in_archive_order":33,"model":"graph2vec","metrics":{"Accuracy":"60.17% ± 6.86%"},"uses_additional_data":false,"paper_date":"2017-07-17","paper":"/paper/graph2vec-learning-distributed","paper_url":"http://arxiv.org/abs/1707.05005v1","paper_title":"graph2vec: Learning Distributed Representations of Graphs","code":"https://github.com/benedekrozemberczki/karateclub","n_code_links":6,"syntology":{"n_ran":2,"n_unverified":0,"n_samples":2,"n_pointer_only_licence":2}},{"rank_in_archive_order":34,"model":"PATCHY-SAN","metrics":{"Accuracy":"60.00%"},"uses_additional_data":false,"paper_date":"2016-05-17","paper":"/paper/learning-convolutional-neural-networks-for","paper_url":"http://arxiv.org/abs/1605.05273v4","paper_title":"Learning Convolutional Neural Networks for Graphs","code":"https://github.com/tvayer/PSCN","n_code_links":2,"syntology":null},{"rank_in_archive_order":35,"model":"IsoNN","metrics":{"Accuracy":"59.9%"},"uses_additional_data":false,"paper_date":"2019-07-22","paper":"/paper/isonn-isomorphic-neural-network-for-graph","paper_url":"https://arxiv.org/abs/1907.09495v2","paper_title":"IsoNN: Isomorphic Neural Network for Graph Representation Learning and Classification","code":"https://github.com/linmengsysu/IsoNN","n_code_links":2,"syntology":null},{"rank_in_archive_order":36,"model":"TREE-G","metrics":{"Accuracy":"59.1%"},"uses_additional_data":false,"paper_date":"2022-07-06","paper":"/paper/graph-trees-with-attention","paper_url":"https://arxiv.org/abs/2207.02760v5","paper_title":"TREE-G: Decision Trees Contesting Graph Neural Networks","code":"https://github.com/mayabechlerspeicher/tree-g","n_code_links":1,"syntology":{"n_ran":9,"n_unverified":1,"n_samples":10,"n_pointer_only_licence":0}},{"rank_in_archive_order":37,"model":"SPI-GCN","metrics":{"Accuracy":"56.41%"},"uses_additional_data":false,"paper_date":"2019-04-08","paper":"/paper/spi-gcn-a-simple-permutation-invariant-graph","paper_url":"https://hal.archives-ouvertes.fr/hal-02093451/","paper_title":"SPI-GCN: A Simple Permutation-Invariant Graph Convolutional Network","code":null,"n_code_links":0,"syntology":null}],"since_archive":{"claim":"Results that newer papers report for their own method, placed here by Syntology. A model pointed at the cell in the paper's own table; the number was read from that cell and checked against this leaderboard's metric, dataset, split and scale; an independent check that saw this leaderboard's other rows and every other leaderboard on the same dataset accepted it. Not reviewed by the paper's authors or by the archive's editors, and not ranked against the archive rows.","extraction_file_present":true,"measurement":{"test_papers":883,"papers_with_output":881,"judged_true":108,"judged":110,"wilson95_lower":0.9361,"measured_on":"2026-09-24","frozen_commit":"0e3de0df94"},"measurement_note":"blind adjudication of accepted entries on a held-out split of archive papers, rules frozen before the test","coverage":{"sentence":"Syntology has checked 6,885 of the 9,623 papers on this site that are newer than the archive; results from the others appear after they are checked.","complete":false,"papers_newer_than_archive":9623,"papers_checked":6885,"papers_extracted_not_yet_verified":0,"boards_without_verdict":2,"papers_not_yet_extracted":2737},"order":"newest first by month (arXiv date, else the arXiv-id month), then arXiv id descending","columns":[],"entries":[]},"syntology":{"read_at":"2026-09-25T09:33:49+00:00","claim":"Per row: N of M harvested code samples from that row's paper executed on a synthesized fixture; the other M-N are unverified. Not a reproduction of the row's number; not a correctness claim. n_pointer_only_licence counts samples the site points at rather than redistributes (a licence axis, independent of ran/unverified).","rows_with_graph_line":11,"rows_with_any_sample_ran":9,"distinct_papers_with_graph_line":10,"distinct_papers_with_any_sample_ran":8,"samples_over_distinct_papers":{"n_ran":25,"n_unverified":31,"n_samples":56,"n_pointer_only_licence":8,"note":"each paper (arXiv id) counted once, however many rows it is behind; this is the page-level figure"},"samples_row_weighted":{"n_ran":29,"n_unverified":31,"n_samples":60,"n_pointer_only_licence":8,"note":"row-weighted: a paper behind several rows is counted once per row; inflated relative to samples_over_distinct_papers by design, kept for readers summing the per-row syntology blocks"}}}