Node Classification On Cornell

Metrics

Accuracy

Results

Performance results of various models on this benchmark

Model Name	Accuracy	Paper Title	Repository
CT-Layer	69.04	DiffWire: Inductive Graph Rewiring via the Lovász Bound	-
ACM-GCN++	85.68 ± 5.8	Revisiting Heterophily For Graph Neural Networks	-
RDGNN-I	92.72 ± 5.88	Graph Neural Reaction Diffusion Models	-
Gen-NSD	85.68 ± 6.51	Neural Sheaf Diffusion: A Topological Perspective on Heterophily and Oversmoothing in GNNs	-
SADE-GCN	86.21±5.59	Self-attention Dual Embedding for Graphs with Heterophily	-
Geom-GCN-I	56.76	Geom-GCN: Geometric Graph Convolutional Networks	-
GREET+CausalMP	68.23±2.90	Heterophilic Graph Neural Networks Optimization with Causal Message-passing	-
UniG-Encoder	86.75±6.56	UniG-Encoder: A Universal Feature Encoder for Graph and Hypergraph Node Classification	-
Geom-GCN-S	55.68	Geom-GCN: Geometric Graph Convolutional Networks	-
PathNet	-	Beyond Homophily: Structure-aware Path Aggregation Graph Neural Network
GloGNN	83.51±4.26	Finding Global Homophily in Graph Neural Networks When Meeting Heterophily	-
H2GCN-1	78.11 ± 6.68	Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs	-
ACM-SGC-1	82.43 ± 5.44	Revisiting Heterophily For Graph Neural Networks	-
GPRGCN	78.11 ± 6.55	Adaptive Universal Generalized PageRank Graph Neural Network	-
DeltaGNN - control + DC	75.67±1.91	DeltaGNN: Graph Neural Network with Information Flow Control	-
ACM-SGC-2	82.43 ± 5.44	Revisiting Heterophily For Graph Neural Networks	-
ACMII-GCN	85.95 ± 5.64	Revisiting Heterophily For Graph Neural Networks	-
GRADE-GAT	83.3±7.0	Graph Neural Aggregation-diffusion with Metastability	-
CNMPGNN	82.38 ± 6.13	CN-Motifs Perceptive Graph Neural Networks	-
ACMII-GCN++	86.49 ± 6.73	Revisiting Heterophily For Graph Neural Networks	-

0 of 60 row(s) selected.