{"url":"/task/node-classification-on-non-homophilic","name":"Node Classification on Non-Homophilic (Heterophilic) Graphs","slug":"node-classification-on-non-homophilic","description_markdown":"There exists a non-trivial set of graphs where graph-aware models underperform their corresponding graph-agnostic models, e.g. SGC and GCN underperform MLP with 1 layer and 2 layers. Although still controversial, people believe the performance degradation results from heterophily, i.e. there exist much more inter-class edges than inner-class edges. This task aims to evaluate models designed for non-homophilic (heterophilic) datasets.","categories":[{"name":"Graphs","url":"/area/graphs"}],"source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","slug_source":"archive_url"},"counts":{"papers_tagged":29,"papers_with_code":29,"benchmarks":15,"benchmark_tables_in_archive":15,"benchmark_tables_shown":15,"benchmark_tables_withheld_as_spam":0,"benchmark_definition":"a leaderboard table with at least one row; benchmark_tables_shown also counts the zero-row tables; benchmark_tables_in_archive adds the tables withheld as spam","datasets":15,"subtasks":0,"parent_tasks":1},"benchmarks":[{"leaderboard":"/sota/node-classification-on-non-homophilic","slug":"node-classification-on-non-homophilic","dataset":"Cornell (60%/20%/20% random splits)","dataset_url":"/dataset/cornell-60-20-20-random-splits","rows_in_archive":33,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ACMII-GCN","paper_title":"Revisiting Heterophily For Graph Neural Networks","paper_url":"/paper/revisiting-heterophily-for-graph-neural","paper_date":"2022-10-14","arxiv_id":"2210.07606","code_links":[{"title":"SitaoLuan/ACM-GNN","url":"https://github.com/SitaoLuan/ACM-GNN"}],"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-1","slug":"node-classification-on-non-homophilic-1","dataset":"Wisconsin(60%/20%/20% random splits)","dataset_url":"/dataset/wisconsin-60-20-20-random-splits-1","rows_in_archive":32,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ACM-GCN++","paper_title":"Revisiting Heterophily For Graph Neural Networks","paper_url":"/paper/revisiting-heterophily-for-graph-neural","paper_date":"2022-10-14","arxiv_id":"2210.07606","code_links":[{"title":"SitaoLuan/ACM-GNN","url":"https://github.com/SitaoLuan/ACM-GNN"}],"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-2","slug":"node-classification-on-non-homophilic-2","dataset":"Texas(60%/20%/20% random splits)","dataset_url":"/dataset/texas-60-20-20-random-splits-1","rows_in_archive":32,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ACM-GCN++","paper_title":"Revisiting Heterophily For Graph Neural Networks","paper_url":"/paper/revisiting-heterophily-for-graph-neural","paper_date":"2022-10-14","arxiv_id":"2210.07606","code_links":[{"title":"SitaoLuan/ACM-GNN","url":"https://github.com/SitaoLuan/ACM-GNN"}],"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-4","slug":"node-classification-on-non-homophilic-4","dataset":"Chameleon(60%/20%/20% random splits)","dataset_url":"/dataset/chameleon-60-20-20-random-splits-1","rows_in_archive":32,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ACM-GCN+","paper_title":"Revisiting Heterophily For Graph Neural Networks","paper_url":"/paper/revisiting-heterophily-for-graph-neural","paper_date":"2022-10-14","arxiv_id":"2210.07606","code_links":[{"title":"SitaoLuan/ACM-GNN","url":"https://github.com/SitaoLuan/ACM-GNN"}],"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-11","slug":"node-classification-on-non-homophilic-11","dataset":"Chameleon (48%/32%/20% fixed splits)","dataset_url":"/dataset/chameleon-48-32-20-fixed-splits","rows_in_archive":29,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ScaleNet","paper_title":"Scale Invariance of Graph Neural Networks","paper_url":"/paper/scale-invariance-of-graph-neural-networks","paper_date":"2024-11-28","arxiv_id":"2411.19392","code_links":[{"title":"qin87/scalenet","url":"https://github.com/qin87/scalenet"}],"syntology":null}},{"leaderboard":"/sota/node-classification-on-non-homophilic-12","slug":"node-classification-on-non-homophilic-12","dataset":"Squirrel (48%/32%/20% fixed splits)","dataset_url":"/dataset/squirrel-48-32-20-fixed-splits","rows_in_archive":29,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ScaleNet","paper_title":"Scale Invariance of Graph Neural Networks","paper_url":"/paper/scale-invariance-of-graph-neural-networks","paper_date":"2024-11-28","arxiv_id":"2411.19392","code_links":[{"title":"qin87/scalenet","url":"https://github.com/qin87/scalenet"}],"syntology":null}},{"leaderboard":"/sota/node-classification-on-non-homophilic-13","slug":"node-classification-on-non-homophilic-13","dataset":"Penn94","dataset_url":"/dataset/penn94","rows_in_archive":28,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ACM-GCN++","paper_title":"Revisiting Heterophily For Graph Neural Networks","paper_url":"/paper/revisiting-heterophily-for-graph-neural","paper_date":"2022-10-14","arxiv_id":"2210.07606","code_links":[{"title":"SitaoLuan/ACM-GNN","url":"https://github.com/SitaoLuan/ACM-GNN"}],"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-6","slug":"node-classification-on-non-homophilic-6","dataset":"Deezer-Europe","dataset_url":"/dataset/deezer-europe-1","rows_in_archive":28,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ACMII-GCN+++","paper_title":"Revisiting Heterophily For Graph Neural Networks","paper_url":"/paper/revisiting-heterophily-for-graph-neural","paper_date":"2022-10-14","arxiv_id":"2210.07606","code_links":[{"title":"SitaoLuan/ACM-GNN","url":"https://github.com/SitaoLuan/ACM-GNN"}],"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-7","slug":"node-classification-on-non-homophilic-7","dataset":"Cornell (48%/32%/20% fixed splits)","dataset_url":"/dataset/cornell-48-32-20-fixed-splits","rows_in_archive":27,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"Diag-NSD","paper_title":"Neural Sheaf Diffusion: A Topological Perspective on Heterophily and Oversmoothing in GNNs","paper_url":"/paper/neural-sheaf-diffusion-a-topological","paper_date":"2022-02-09","arxiv_id":"2202.04579","code_links":[{"title":"twitter-research/neural-sheaf-diffusion","url":"https://github.com/twitter-research/neural-sheaf-diffusion"}],"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-10","slug":"node-classification-on-non-homophilic-10","dataset":"Film(48%/32%/20% fixed splits)","dataset_url":"/dataset/film-48-32-20-fixed-splits","rows_in_archive":26,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"NLMLP","paper_title":"Non-Local Graph Neural Networks","paper_url":"/paper/non-local-graph-neural-networks","paper_date":"2020-05-29","arxiv_id":"2005.14612","code_links":[{"title":"divelab/Non-Local-GNN","url":"https://github.com/divelab/Non-Local-GNN"}],"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":3}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-14","slug":"node-classification-on-non-homophilic-14","dataset":"genius","dataset_url":"/dataset/genius","rows_in_archive":26,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ClenshawGCN","paper_title":"Clenshaw Graph Neural Networks","paper_url":"/paper/clenshaw-graph-neural-networks","paper_date":"2022-10-29","arxiv_id":"2210.16508","code_links":[{"title":"yuziGuo/ClenshawGNN","url":"https://github.com/yuziGuo/ClenshawGNN"}],"syntology":null}},{"leaderboard":"/sota/node-classification-on-non-homophilic-15","slug":"node-classification-on-non-homophilic-15","dataset":"twitch-gamers","dataset_url":"/dataset/twitch-gamers","rows_in_archive":26,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"GESN","paper_title":"Addressing Heterophily in Node Classification with Graph Echo State Networks","paper_url":"/paper/addressing-heterophily-in-node-classification","paper_date":"2023-05-14","arxiv_id":"2305.08233","code_links":[{"title":"dtortorella/addressing-heterophily-gesn","url":"https://github.com/dtortorella/addressing-heterophily-gesn"}],"syntology":null}},{"leaderboard":"/sota/node-classification-on-non-homophilic-8","slug":"node-classification-on-non-homophilic-8","dataset":"Wisconsin (48%/32%/20% fixed splits)","dataset_url":"/dataset/wisconsin-48-32-20-fixed-splits","rows_in_archive":26,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"O(d)-NSD","paper_title":"Neural Sheaf Diffusion: A Topological Perspective on Heterophily and Oversmoothing in GNNs","paper_url":"/paper/neural-sheaf-diffusion-a-topological","paper_date":"2022-02-09","arxiv_id":"2202.04579","code_links":[{"title":"twitter-research/neural-sheaf-diffusion","url":"https://github.com/twitter-research/neural-sheaf-diffusion"}],"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-9","slug":"node-classification-on-non-homophilic-9","dataset":"Texas (48%/32%/20% fixed splits)","dataset_url":"/dataset/texas-48-32-20-fixed-splits","rows_in_archive":26,"metrics":["1:1 Accuracy"],"first_row_in_archive_order":{"model":"ACM-GCN+","paper_title":"Revisiting Heterophily For Graph Neural Networks","paper_url":"/paper/revisiting-heterophily-for-graph-neural","paper_date":"2022-10-14","arxiv_id":"2210.07606","code_links":[{"title":"SitaoLuan/ACM-GNN","url":"https://github.com/SitaoLuan/ACM-GNN"}],"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}}},{"leaderboard":"/sota/node-classification-on-non-homophilic-16","slug":"node-classification-on-non-homophilic-16","dataset":"Pubmed","dataset_url":"/dataset/pubmed","rows_in_archive":1,"metrics":["F1-Score","NMI"],"first_row_in_archive_order":{"model":"GAT","paper_title":"Graph Attention Networks","paper_url":"/paper/graph-attention-networks","paper_date":"2017-10-30","arxiv_id":"1710.10903","code_links":[{"title":"labmlai/annotated_deep_learning_paper_implementations","url":"https://github.com/labmlai/annotated_deep_learning_paper_implementations"},{"title":"dmlc/dgl","url":"https://github.com/dmlc/dgl/tree/master/examples/pytorch/gat"},{"title":"dmlc/dgl","url":"https://github.com/dmlc/dgl/tree/master/examples/tensorflow/gat"},{"title":"dmlc/dgl","url":"https://github.com/dmlc/dgl/tree/master/examples/mxnet/gat"},{"title":"PetarV-/GAT","url":"https://github.com/PetarV-/GAT"},{"title":"Diego999/pyGAT","url":"https://github.com/Diego999/pyGAT"},{"title":"gordicaleksa/pytorch-GAT","url":"https://github.com/gordicaleksa/pytorch-GAT"},{"title":"shenweichen/GraphNeuralNetwork","url":"https://github.com/shenweichen/GraphNeuralNetwork"},{"title":"HazyResearch/hgcn","url":"https://github.com/HazyResearch/hgcn"},{"title":"GraphSAINT/GraphSAINT","url":"https://github.com/GraphSAINT/GraphSAINT"},{"title":"danielegrattarola/keras-gat","url":"https://github.com/danielegrattarola/keras-gat"},{"title":"lukecavabarrett/pna","url":"https://github.com/lukecavabarrett/pna"},{"title":"meliketoy/graph-cnn.pytorch","url":"https://github.com/meliketoy/graph-cnn.pytorch"},{"title":"mindspore-ai/models","url":"https://github.com/mindspore-ai/models/tree/master/official/gnn/gat"},{"title":"Kaimaoge/IGNNK","url":"https://github.com/Kaimaoge/IGNNK"},{"title":"IllinoisGraphBenchmark/IGB-Datasets","url":"https://github.com/IllinoisGraphBenchmark/IGB-Datasets"},{"title":"arangoml/fastgraphml","url":"https://github.com/arangoml/fastgraphml"},{"title":"zhao-tong/GNNs-easy-to-use","url":"https://github.com/zhao-tong/GNNs-easy-to-use"},{"title":"davidpicard/homm","url":"https://github.com/davidpicard/homm"},{"title":"alexOarga/haiku-geometric","url":"https://github.com/alexOarga/haiku-geometric"},{"title":"snowkylin/gnn","url":"https://github.com/snowkylin/gnn"},{"title":"akash18tripathi/MAGNET-Multi-Label-Text-Classi-cation-using-Attention-based-Graph-Neural-Network","url":"https://github.com/akash18tripathi/MAGNET-Multi-Label-Text-Classi-cation-using-Attention-based-Graph-Neural-Network"},{"title":"fanzhenliu/dagad","url":"https://github.com/fanzhenliu/dagad"},{"title":"weiyangfb/PyTorchSparseGAT","url":"https://github.com/weiyangfb/PyTorchSparseGAT"},{"title":"HeapHop30/graph-attention-nets","url":"https://github.com/HeapHop30/graph-attention-nets"},{"title":"marblet/GNN_models_pytorch_geometric","url":"https://github.com/marblet/GNN_models_pytorch_geometric"},{"title":"marble0117/GNN_models_pytorch_geometric","url":"https://github.com/marble0117/GNN_models_pytorch_geometric"},{"title":"marble0117/GNN_models_pytorch","url":"https://github.com/marble0117/GNN_models_pytorch"},{"title":"zxhhh97/ABot","url":"https://github.com/zxhhh97/ABot"},{"title":"ds4dm/sGat","url":"https://github.com/ds4dm/sGat"},{"title":"ds4dm/sparse-gcn","url":"https://github.com/ds4dm/sparse-gcn"},{"title":"isotlaboratory/ml4vrp","url":"https://github.com/isotlaboratory/ml4vrp"},{"title":"marblet/gat-pytorch","url":"https://github.com/marblet/gat-pytorch"},{"title":"ydzhang-stormstout/HAT","url":"https://github.com/ydzhang-stormstout/HAT"},{"title":"gcucurull/jax-gat","url":"https://github.com/gcucurull/jax-gat"},{"title":"calciver/Graph-Attention-Networks","url":"https://github.com/calciver/Graph-Attention-Networks"},{"title":"shi27feng/transformers.satisfy","url":"https://github.com/shi27feng/transformers.satisfy"},{"title":"pengsl-lab/mssl","url":"https://github.com/pengsl-lab/mssl"},{"title":"pengsl-lab/mssl2drug","url":"https://github.com/pengsl-lab/mssl2drug"},{"title":"nnaakkaaii/g2-MLP","url":"https://github.com/nnaakkaaii/g2-MLP"},{"title":"gmum/umwpl2021","url":"https://github.com/gmum/umwpl2021"},{"title":"taishan1994/pytorch_gat","url":"https://github.com/taishan1994/pytorch_gat"},{"title":"wan2000/mc-mot","url":"https://github.com/wan2000/mc-mot"},{"title":"basiralab/RG-Select","url":"https://github.com/basiralab/RG-Select"},{"title":"dmeoli/neuro-sat","url":"https://github.com/dmeoli/neuro-sat"},{"title":"liu6zijian/simplified-gcn-model","url":"https://github.com/liu6zijian/simplified-gcn-model"},{"title":"tatp22/pytorch-fast-GAT","url":"https://github.com/tatp22/pytorch-fast-GAT"},{"title":"noahtren/Graph-Attention-Networks-TensorFlow-2","url":"https://github.com/noahtren/Graph-Attention-Networks-TensorFlow-2"},{"title":"yururyuru00/GAT_pyG","url":"https://github.com/yururyuru00/GAT_pyG"},{"title":"qema/orca-py","url":"https://github.com/qema/orca-py"},{"title":"zhanglab-aim/cancer-net","url":"https://github.com/zhanglab-aim/cancer-net"},{"title":"gayanku/FDGATII","url":"https://github.com/gayanku/FDGATII"},{"title":"Yindong-Zhang/myGAT","url":"https://github.com/Yindong-Zhang/myGAT"},{"title":"giuseppefutia/link-prediction-code","url":"https://github.com/giuseppefutia/link-prediction-code"},{"title":"Aveek-Saha/Graph-Attention-Net","url":"https://github.com/Aveek-Saha/Graph-Attention-Net"},{"title":"BIG-S2/keras-gnm","url":"https://github.com/BIG-S2/keras-gnm"},{"title":"mlzxzhou/keras-gnm","url":"https://github.com/mlzxzhou/keras-gnm"},{"title":"basiralab/Reproducible-Generative-Learning","url":"https://github.com/basiralab/Reproducible-Generative-Learning"},{"title":"abhilash1910/GraphAttentionNetworks","url":"https://github.com/abhilash1910/GraphAttentionNetworks"},{"title":"Kristof-Neys/GAT_Neo4j_DGL","url":"https://github.com/Kristof-Neys/GAT_Neo4j_DGL"},{"title":"ak24watch/Graph-Deeplearning-GAT","url":"https://github.com/ak24watch/Graph-Deeplearning-GAT"},{"title":"Negus25/GAT-Mindspore","url":"https://github.com/Negus25/GAT-Mindspore"},{"title":"ChengyuSun/gat","url":"https://github.com/ChengyuSun/gat"},{"title":"joaopedromattos/gnee","url":"https://github.com/joaopedromattos/gnee"},{"title":"mitya8128/experiments_notes","url":"https://github.com/mitya8128/experiments_notes"},{"title":"4pygmalion/GAT_tensorflow","url":"https://github.com/4pygmalion/GAT_tensorflow"},{"title":"vailatuts/SSFG-regularization","url":"https://github.com/vailatuts/SSFG-regularization"},{"title":"handasontam/GAT-with-edgewise-attention","url":"https://github.com/handasontam/GAT-with-edgewise-attention"},{"title":"Mind23-2/MindCode-4","url":"https://github.com/Mind23-2/MindCode-4/tree/main/gat"},{"title":"AngusMonroe/GAT-pytorch","url":"https://github.com/AngusMonroe/GAT-pytorch"},{"title":"fongyk/graph-attention","url":"https://github.com/fongyk/graph-attention"},{"title":"404nofound/gat","url":"https://github.com/404nofound/gat"},{"title":"ColdenChan/GAT_D","url":"https://github.com/ColdenChan/GAT_D"},{"title":"MehmetBarutcu/Streaming-Algorithm-for-Monotone-k-Submodular-Maximization-with-Cardinality-Constraints","url":"https://github.com/MehmetBarutcu/Streaming-Algorithm-for-Monotone-k-Submodular-Maximization-with-Cardinality-Constraints"},{"title":"bowu/GraphSearch","url":"https://github.com/bowu/GraphSearch"},{"title":"Messham87/COVID_GAT","url":"https://github.com/Messham87/COVID_GAT"},{"title":"zhangbo2008/GAT_network","url":"https://github.com/zhangbo2008/GAT_network"},{"title":"blueberryc/pyGAT","url":"https://github.com/blueberryc/pyGAT"},{"title":"anish-lu-yihe/SVRT-by-GAT","url":"https://github.com/anish-lu-yihe/SVRT-by-GAT"},{"title":"WantingZhao/my_GAT","url":"https://github.com/WantingZhao/my_GAT"},{"title":"dzb1998/pyGAT","url":"https://github.com/dzb1998/pyGAT"},{"title":"subercui/pyGConvAT","url":"https://github.com/subercui/pyGConvAT"},{"title":"YunseobShin/wiki_GAT","url":"https://github.com/YunseobShin/wiki_GAT"},{"title":"MindSpore-scientific-2/code-4","url":"https://github.com/MindSpore-scientific-2/code-4/tree/main/GR-HSTU"},{"title":"snownus/COOP","url":"https://github.com/snownus/COOP"},{"title":"ChengSashankh/trying-GAT","url":"https://github.com/ChengSashankh/trying-GAT"},{"title":"Anak2016/GAT","url":"https://github.com/Anak2016/GAT"},{"title":"whut2962575697/gat_sementic_segmentation","url":"https://github.com/whut2962575697/gat_sementic_segmentation"},{"title":"Anou9531/GAT","url":"https://github.com/Anou9531/GAT"},{"title":"PumpkinYing/GAT","url":"https://github.com/PumpkinYing/GAT"},{"title":"galkampel/HyperNetworks","url":"https://github.com/galkampel/HyperNetworks"},{"title":"tlmakinen/GAT-noise","url":"https://github.com/tlmakinen/GAT-noise"},{"title":"TyngJiunKuo/deep-learning-project","url":"https://github.com/TyngJiunKuo/deep-learning-project"}],"syntology":{"n":106,"n_ran":50,"n_unverified":56,"n_pointer_only":43}}}],"datasets":[{"url":"/dataset/pubmed","name":"Pubmed","full_name":"","num_papers_in_archive":1236},{"url":"/dataset/penn94","name":"Penn94","full_name":"","num_papers_in_archive":60},{"url":"/dataset/genius","name":"genius","full_name":"","num_papers_in_archive":37},{"url":"/dataset/twitch-gamers","name":"twitch-gamers","full_name":"","num_papers_in_archive":24},{"url":"/dataset/chameleon-48-32-20-fixed-splits","name":"Chameleon (48%/32%/20% fixed splits)","full_name":"","num_papers_in_archive":20},{"url":"/dataset/deezer-europe-1","name":"Deezer-Europe","full_name":"","num_papers_in_archive":19},{"url":"/dataset/squirrel-48-32-20-fixed-splits","name":"Squirrel (48%/32%/20% fixed splits)","full_name":"","num_papers_in_archive":19},{"url":"/dataset/wisconsin-60-20-20-random-splits-1","name":"Wisconsin(60%/20%/20% random splits)","full_name":"","num_papers_in_archive":18},{"url":"/dataset/chameleon-60-20-20-random-splits-1","name":"Chameleon(60%/20%/20% random splits)","full_name":"","num_papers_in_archive":17},{"url":"/dataset/cornell-48-32-20-fixed-splits","name":"Cornell (48%/32%/20% fixed splits)","full_name":"","num_papers_in_archive":16},{"url":"/dataset/cornell-60-20-20-random-splits","name":"Cornell (60%/20%/20% random splits)","full_name":"","num_papers_in_archive":16},{"url":"/dataset/texas-60-20-20-random-splits-1","name":"Texas(60%/20%/20% random splits)","full_name":"","num_papers_in_archive":16},{"url":"/dataset/wisconsin-48-32-20-fixed-splits","name":"Wisconsin (48%/32%/20% fixed splits)","full_name":"","num_papers_in_archive":15},{"url":"/dataset/film-48-32-20-fixed-splits","name":"Film(48%/32%/20% fixed splits)","full_name":"","num_papers_in_archive":14},{"url":"/dataset/texas-48-32-20-fixed-splits","name":"Texas (48%/32%/20% fixed splits)","full_name":"","num_papers_in_archive":14}],"subtasks":[],"parent_tasks":[{"url":"/task/node-classification","name":"Node Classification"}],"papers":{"order":"repositories listed in the archive (desc), then date (desc); the archive holds no stars","population":"papers tagged with this task that list at least one repository in the archive","shown":29,"of":29,"tagged_in_all":29,"items":[{"url":"/paper/graph-attention-networks","title":"Graph Attention Networks","date":"2017-10-30","arxiv_id":"1710.10903","repositories_listed":93,"syntology":{"n":106,"n_ran":50,"n_unverified":56,"n_pointer_only":43}},{"url":"/paper/semi-supervised-classification-with-graph","title":"Semi-Supervised Classification with Graph Convolutional Networks","date":"2016-09-09","arxiv_id":"1609.02907","repositories_listed":55,"syntology":{"n":58,"n_ran":31,"n_unverified":27,"n_pointer_only":22}},{"url":"/paper/inductive-representation-learning-on-large","title":"Inductive Representation Learning on Large Graphs","date":"2017-06-07","arxiv_id":"1706.02216","repositories_listed":20,"syntology":{"n":5,"n_ran":3,"n_unverified":2,"n_pointer_only":5}},{"url":"/paper/combining-label-propagation-and-simple-models-1","title":"Combining Label Propagation and Simple Models Out-performs Graph Neural Networks","date":"2020-10-27","arxiv_id":"2010.13993","repositories_listed":7,"syntology":{"n":5,"n_ran":2,"n_unverified":3,"n_pointer_only":2}},{"url":"/paper/simplifying-graph-convolutional-networks","title":"Simplifying Graph Convolutional Networks","date":"2019-02-19","arxiv_id":"1902.07153","repositories_listed":7,"syntology":{"n":8,"n_ran":3,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/large-scale-learning-on-non-homophilous","title":"Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methods","date":"2021-10-27","arxiv_id":"2110.14446","repositories_listed":5,"syntology":{"n":21,"n_ran":7,"n_unverified":14,"n_pointer_only":4}},{"url":"/paper/predict-then-propagate-graph-neural-networks","title":"Predict then Propagate: Graph Neural Networks meet Personalized PageRank","date":"2018-10-14","arxiv_id":"1810.05997","repositories_listed":5,"syntology":{"n":12,"n_ran":1,"n_unverified":11,"n_pointer_only":0}},{"url":"/paper/simple-and-deep-graph-convolutional-networks-1","title":"Simple and Deep Graph Convolutional Networks","date":"2020-07-04","arxiv_id":"2007.02133","repositories_listed":4,"syntology":{"n":7,"n_ran":5,"n_unverified":2,"n_pointer_only":4}},{"url":"/paper/generalizing-graph-neural-networks-beyond","title":"Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs","date":"2020-06-20","arxiv_id":"2006.11468","repositories_listed":4,"syntology":{"n":21,"n_ran":13,"n_unverified":8,"n_pointer_only":6}},{"url":"/paper/geom-gcn-geometric-graph-convolutional-1","title":"Geom-GCN: Geometric Graph Convolutional Networks","date":"2020-02-13","arxiv_id":"2002.05287","repositories_listed":4,"syntology":{"n":7,"n_ran":7,"n_unverified":0,"n_pointer_only":7}},{"url":"/paper/mixhop-higher-order-graph-convolution","title":"MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing","date":"2019-04-30","arxiv_id":"1905.00067","repositories_listed":3,"syntology":null},{"url":"/paper/scale-invariance-of-graph-neural-networks","title":"Scale Invariance of Graph Neural Networks","date":"2024-11-28","arxiv_id":"2411.19392","repositories_listed":1,"syntology":null},{"url":"/paper/edge-directionality-improves-learning-on","title":"Edge Directionality Improves Learning on Heterophilic Graphs","date":"2023-05-17","arxiv_id":"2305.10498","repositories_listed":1,"syntology":{"n":5,"n_ran":3,"n_unverified":2,"n_pointer_only":0}},{"url":"/paper/addressing-heterophily-in-node-classification","title":"Addressing Heterophily in Node Classification with Graph Echo State Networks","date":"2023-05-14","arxiv_id":"2305.08233","repositories_listed":1,"syntology":null},{"url":"/paper/gcnh-a-simple-method-for-representation","title":"GCNH: A Simple Method For Representation Learning On Heterophilous Graphs","date":"2023-04-21","arxiv_id":"2304.10896","repositories_listed":1,"syntology":null},{"url":"/paper/graph-neural-networks-with-learnable-and","title":"Graph Neural Networks with Learnable and Optimal Polynomial Bases","date":"2023-02-24","arxiv_id":"2302.12432","repositories_listed":1,"syntology":null},{"url":"/paper/clenshaw-graph-neural-networks","title":"Clenshaw Graph Neural Networks","date":"2022-10-29","arxiv_id":"2210.16508","repositories_listed":1,"syntology":null},{"url":"/paper/revisiting-heterophily-for-graph-neural","title":"Revisiting Heterophily For Graph Neural Networks","date":"2022-10-14","arxiv_id":"2210.07606","repositories_listed":1,"syntology":{"n":7,"n_ran":2,"n_unverified":5,"n_pointer_only":0}},{"url":"/paper/finding-global-homophily-in-graph-neural","title":"Finding Global Homophily in Graph Neural Networks When Meeting Heterophily","date":"2022-05-15","arxiv_id":"2205.07308","repositories_listed":1,"syntology":null},{"url":"/paper/neural-sheaf-diffusion-a-topological","title":"Neural Sheaf Diffusion: A Topological Perspective on Heterophily and Oversmoothing in GNNs","date":"2022-02-09","arxiv_id":"2202.04579","repositories_listed":1,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":0}},{"url":"/paper/deformable-graph-convolutional-networks","title":"Deformable Graph Convolutional Networks","date":"2021-12-29","arxiv_id":"2112.14438","repositories_listed":1,"syntology":{"n":2,"n_ran":2,"n_unverified":0,"n_pointer_only":0}},{"url":"/paper/bernnet-learning-arbitrary-graph-spectral","title":"BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation","date":"2021-06-21","arxiv_id":"2106.10994","repositories_listed":1,"syntology":{"n":1,"n_ran":0,"n_unverified":1,"n_pointer_only":1}},{"url":"/paper/breaking-the-limit-of-graph-neural-networks","title":"Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns","date":"2021-06-11","arxiv_id":"2106.06586","repositories_listed":1,"syntology":{"n":12,"n_ran":1,"n_unverified":11,"n_pointer_only":0}},{"url":"/paper/new-benchmarks-for-learning-on-non","title":"New Benchmarks for Learning on Non-Homophilous Graphs","date":"2021-04-03","arxiv_id":"2104.01404","repositories_listed":1,"syntology":null},{"url":"/paper/two-sides-of-the-same-coin-heterophily-and","title":"Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks","date":"2021-02-12","arxiv_id":"2102.06462","repositories_listed":1,"syntology":null},{"url":"/paper/beyond-low-frequency-information-in-graph","title":"Beyond Low-frequency Information in Graph Convolutional Networks","date":"2021-01-04","arxiv_id":"2101.00797","repositories_listed":1,"syntology":null},{"url":"/paper/joint-adaptive-feature-smoothing-and-topology","title":"Adaptive Universal Generalized PageRank Graph Neural Network","date":"2020-06-14","arxiv_id":"2006.07988","repositories_listed":1,"syntology":{"n":2,"n_ran":0,"n_unverified":2,"n_pointer_only":2}},{"url":"/paper/non-local-graph-neural-networks","title":"Non-Local Graph Neural Networks","date":"2020-05-29","arxiv_id":"2005.14612","repositories_listed":1,"syntology":{"n":3,"n_ran":3,"n_unverified":0,"n_pointer_only":3}},{"url":"/paper/break-the-ceiling-stronger-multi-scale-deep","title":"Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks","date":"2019-06-05","arxiv_id":"1906.02174","repositories_listed":1,"syntology":{"n":9,"n_ran":1,"n_unverified":8,"n_pointer_only":0}}],"syntology_records":19,"syntology_note":"a paper without a record is not a recorded non-run: it may lack an arXiv id or simply be absent from the graph layer"},"description_links":{"kept":0,"unwrapped_to_text":0,"bare_urls_linked":0,"relative_images_dropped":0,"rule":"internal links are kept only when the target slug exists in the catalog"},"syntology":{"read_at":"2026-09-24T18:15:14+00:00","claim":"Per-sample execution status on synthesized fixtures ('ran N of M samples'); not a correctness claim and not a ranking signal.","status_vocabulary":{"ran_honours":"ran, honoured the contract we drafted","ran_violates":"ran, violated the contract we drafted","ran_draft_wrong":"ran; our contract draft was wrong, not the code","ran_fixture":"ran; our fixture could not drive it","ran":"ran on a synthesized input","unverified":"unverified (harvested, no recorded run)"}},"not_shown":{"libraries":"the archive has no per-task library table","trend_sparklines":"the Trend column of the benchmarks table was a rendered image; it is not in the archive","social_and_latest_sorts":"stars and social signals are not in the archive"}}