Papers › Clarify Confused Nodes via Separated Learning

Clarify Confused Nodes via Separated Learning

4 Jun 2023arXiv:2306.02285archive 2025-07-28

Jiajun Zhou, Shengbo Gong, Xuanze Chen, Chenxuan Xie, Shanqing Yu, Qi Xuan, Xiaoniu Yang

Graph neural networks (GNNs) have achieved remarkable advances in graph-oriented tasks. However, real-world graphs invariably contain a certain proportion of heterophilous nodes, challenging the homophily assumption of traditional GNNs and hindering their performance. Most existing studies continue to design generic models with shared weights between heterophilous and homophilous nodes. Despite the incorporation of high-order messages or multi-channel architectures, these efforts often fall short. A minority of studies attempt to train different node groups separately but suffer from inappropriate separation metrics and low efficiency. In this paper, we first propose a new metric, termed Neighborhood Confusion (NC), to facilitate a more reliable separation of nodes. We observe that node groups with different levels of NC values exhibit certain differences in intra-group accuracy and visualized embeddings. These pave the way for Neighborhood Confusion-guided Graph Convolutional Network (NCGCN), in which nodes are grouped by their NC values and accept intra-group weight sharing and message passing. Extensive experiments on both homophilous and heterophilous benchmarks demonstrate that our framework can effectively separate nodes and yield significant performance improvement compared to the latest methods. The source code will be available in https://github.com/GISec-Team/NCGNN.

PaperPDFCode

Code

GISec-Team/NCGNN officialmentioned in papermentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Node Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Node Classification AMZ Computers NCGCN Accuracy 90.81 ± 0.46 #1 of 5 Archive leaderboard report
Node Classification AMZ Computers NCSAGE Accuracy 90.43 ± 0.72 #2 of 5 Archive leaderboard report
Node Classification AMZ Photo NCSAGE Accuracy 95.93 ± 0.36 #1 of 14 Archive leaderboard report
Node Classification AMZ Photo NCGCN Accuracy 95.45 ± 0.45 #3 of 14 Archive leaderboard report
Node Classification Actor NCSAGE Accuracy 43.89 ± 1.33 #1 of 62 Archive leaderboard report
Node Classification Actor NCGCN Accuracy 43.16 ± 1.32 #2 of 62 Archive leaderboard report
Node Classification Chameleon(60%/20%/20% random splits) NCSAGE Accuracy 48.07 ± 3.47 #1 of 2 Archive leaderboard report
Node Classification Chameleon(60%/20%/20% random splits) NCGCN Accuracy 46.84 ± 4.36 #2 of 2 Archive leaderboard report
Node Classification Coauthor CS NCGCN Accuracy 96.64 ± 0.29 #1 of 24 Archive leaderboard report
Node Classification Coauthor CS NCSAGE Accuracy 96.48 ± 0.25 #2 of 24 Archive leaderboard report
Node Classification Coauthor Physics NCSAGE Accuracy 98.69 ± 0.26 #1 of 14 Archive leaderboard report
Node Classification Coauthor Physics NCGCN Accuracy 98.63 ± 0.24 #2 of 14 Archive leaderboard report
Node Classification Cora Full NCSAGE Accuracy 72.58 ± 0.65% #1 of 5 Archive leaderboard report
Node Classification Cora Full-supervised NCGCN Accuracy 73.42 ± 0.58% #8 of 9 Archive leaderboard report
Node Classification Penn94 NCGCN Accuracy 84.74 ± 0.28 #10 of 32 Archive leaderboard report
Node Classification Penn94 NCSAGE Accuracy 81.77 ± 0.71 #17 of 32 Archive leaderboard report
Node Classification Pubmed NCGCN Accuracy 91.64 ± 0.53 #1 of 70 Archive leaderboard report
Node Classification Pubmed NCSAGE Accuracy 91.55 ± 0.38 #2 of 70 Archive leaderboard report
Node Classification Squirrel(60%/20%/20% random splits) NCSAGE Accuracy 54.42 ± 1.41 #1 of 2 Archive leaderboard report
Node Classification Squirrel(60%/20%/20% random splits) NCGCN Accuracy 54.15 ± 1.47 #2 of 2 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections