HyperAIHyperAI

Command Palette

Search for a command to run...

Console

Node Classification On Chameleon

Metrics

Accuracy

Results

Performance results of various models on this benchmark

Paper Title
DJ-GNN80.48±1.46Diffusion-Jump GNNs: Homophiliation via Learnable Metric Filters
FaberNet80.33±1.19HoloNets: Spectral Convolutions do extend to Directed Graphs
Dir-GNN79.71±1.26Edge Directionality Improves Learning on Heterophilic Graphs
CoED79.69±1.35Improving Graph Neural Networks by Learning Continuous Edge Directions
FSGNN (8-hop)78.27±1.28Improving Graph Neural Networks with Simple Architecture Design
FSGNN (3-hop)78.14±1.25Improving Graph Neural Networks with Simple Architecture Design
HLP Concat77.48±0.80Simple Truncated SVD based Model for Node Classification on Heterophilic Graphs
Graph ESN76.2±1.2Beyond Homophily with Graph Echo State Networks
SADE-GCN75.57±1.57Self-attention Dual Embedding for Graphs with Heterophily
M2M-GNN75.20 ± 2.3Sign is Not a Remedy: Multiset-to-Multiset Message Passing for Learning on Heterophilic Graphs
RDGNN-I74.79 ± 2.14Graph Neural Reaction Diffusion Models
ACMII-GCN++74.76 ± 2.2Revisiting Heterophily For Graph Neural Networks
GCNII+DHGR74.57±2.56Make Heterophily Graphs Better Fit GNN: A Graph Rewiring Approach
ACMII-GCN+74.56 ± 2.08Revisiting Heterophily For Graph Neural Networks
UDGNN (GCN)74.53±1.19Universal Deep GNNs: Rethinking Residual Connection in GNNs from a Path Decomposition Perspective for Preventing the Over-smoothing
ACM-GCN+74.47 ± 1.84Revisiting Heterophily For Graph Neural Networks
ACM-GCN++74.41 ± 1.49Revisiting Heterophily For Graph Neural Networks
LW-GCN74.4±1.4Label-Wise Graph Convolutional Network for Heterophilic Graphs
SignGT74.31±1.24SignGT: Signed Attention-based Graph Transformer for Graph Representation Learning
CNMPGNN73.29±1.29CN-Motifs Perceptive Graph Neural Networks
0 of 61 row(s) selected.
Node Classification On Chameleon | SOTA | HyperAI