Papers › Learn from Heterophily: Heterophilous Information-enhanced Graph Neural Network

Learn from Heterophily: Heterophilous Information-enhanced Graph Neural Network

26 Mar 2024arXiv:2403.17351archive 2025-07-28

Yilun Zheng, Jiahao Xu, Lihui Chen

Under circumstances of heterophily, where nodes with different labels tend to be connected based on semantic meanings, Graph Neural Networks (GNNs) often exhibit suboptimal performance. Current studies on graph heterophily mainly focus on aggregation calibration or neighbor extension and address the heterophily issue by utilizing node features or structural information to improve GNN representations. In this paper, we propose and demonstrate that the valuable semantic information inherent in heterophily can be utilized effectively in graph learning by investigating the distribution of neighbors for each individual node within the graph. The theoretical analysis is carried out to demonstrate the efficacy of the idea in enhancing graph learning. Based on this analysis, we propose HiGNN, an innovative approach that constructs an additional new graph structure, that integrates heterophilous information by leveraging node distribution to enhance connectivity between nodes that share similar semantic characteristics. We conduct empirical assessments on node classification tasks using both homophilous and heterophilous benchmark datasets and compare HiGNN to popular GNN baselines and SoTA methods, confirming the effectiveness in improving graph representations. In addition, by incorporating heterophilous information, we demonstrate a notable enhancement in existing GNN-based approaches, and the homophily degree across real-world datasets, thus affirming the efficacy of our approach.

PaperPDFCode

Code

zylMozart/HiGNN officialpytorch 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

Graph LearningGraph Neural NetworkNode Classification

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Node Classification Actor HiGNN Accuracy 37.21 ± 1.35 #28 of 62 Archive leaderboard report
Node Classification Chameleon HiGNN Accuracy 68.86 ± 1.45 #35 of 61 Archive leaderboard report
Node Classification Cornell HiGNN Accuracy 80.00 ± 4.26 #40 of 60 Archive leaderboard report
Node Classification Squirrel HiGNN Accuracy 54.78 ± 1.58 #40 of 59 Archive leaderboard report
Node Classification Texas HiGNN Accuracy 86.22 ± 4.67 #23 of 62 Archive leaderboard report
Node Classification Wisconsin HiGNN Accuracy 85.88 ± 3.18 #41 of 63 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.

Methods

Focus

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