Node Classification On Squirrel

Metrics

Accuracy

Results

Performance results of various models on this benchmark

Model Name	Accuracy	Paper Title	Repository
CoED	75.32±1.82	Improving Graph Neural Networks by Learning Continuous Edge Directions
MixHop	43.80 ± 1.48	MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing
ACMII-GCN++	67.4 ± 2.21	Revisiting Heterophily For Graph Neural Networks
Conn-NSD	45.19±1.57	Sheaf Neural Networks with Connection Laplacians
WRGAT	48.85 ± 0.78	Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns
Gprompt+CausalMP	39.78±0.91	Heterophilic Graph Neural Networks Optimization with Causal Message-passing	-
Ordered GNN	62.44±1.96	Ordered GNN: Ordering Message Passing to Deal with Heterophily and Over-smoothing
CNMPGNN	63.60±1.96	CN-Motifs Perceptive Graph Neural Networks	-
UDGNN (GCN)	-	Universal Deep GNNs: Rethinking Residual Connection in GNNs from a Path Decomposition Perspective for Preventing the Over-smoothing	-
ACMII-GCN+	67.07 ± 1.65	Revisiting Heterophily For Graph Neural Networks
JKNet + Hetero-S (8 layers)	57.83	The Heterophilic Snowflake Hypothesis: Training and Empowering GNNs for Heterophilic Graphs	-
H2GCN-1	28.98 ± 1.97	Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs
FaberNet	76.71±1.92	HoloNets: Spectral Convolutions do extend to Directed Graphs	-
GloGNN	57.54±1.39	Finding Global Homophily in Graph Neural Networks When Meeting Heterophily
ACM-SGC-2	40.02 ± 0.96	Revisiting Heterophily For Graph Neural Networks
ACM-GCN++	67.06 ± 1.66	Revisiting Heterophily For Graph Neural Networks
ADPA	45.2±1.3	Breaking the Entanglement of Homophily and Heterophily in Semi-supervised Node Classification	-
Graph ESN	71.2±1.5	Beyond Homophily with Graph Echo State Networks	-
LW-GCN	62.6±1.6	Label-Wise Graph Convolutional Network for Heterophilic Graphs
HDP	62.07 ± 1.57	Heterophilous Distribution Propagation for Graph Neural Networks	-

0 of 59 row(s) selected.