Node Classification On Chameleon

Métriques

Accuracy

Résultats

Résultats de performance de divers modèles sur ce benchmark

Nom du modèle	Accuracy	Paper Title	Repository
GCNH	71.56±1.86	GCNH: A Simple Method For Representation Learning On Heterophilous Graphs
Dir-GNN	79.71±1.26	Edge Directionality Improves Learning on Heterophilic Graphs
H2GCN-1	52.96 ± 2.09	Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs
GCNII	63.86 ± 3.04	Simple and Deep Graph Convolutional Networks
H2GCN-2	58.38 ± 1.76	Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs
ACM-GCN	69.14 ± 1.91	Revisiting Heterophily For Graph Neural Networks
SADE-GCN	75.57±1.57	Self-attention Dual Embedding for Graphs with Heterophily	-
NLMLP	50.7 ± 2.2	Non-Local Graph Neural Networks
ACM-SGC-1	63.99 ± 1.66	Revisiting Heterophily For Graph Neural Networks
CNMPGNN	73.29±1.29	CN-Motifs Perceptive Graph Neural Networks	-
CATv3-sup	69.9±1.0	CAT: A Causally Graph Attention Network for Trimming Heterophilic Graph
LINKX	68.42 ± 1.38	Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methods
SDRF	42.73±0.15	Understanding over-squashing and bottlenecks on graphs via curvature
LHS	72.31±1.6	Refining Latent Homophilic Structures over Heterophilic Graphs for Robust Graph Convolution Networks	-
NLGCN	70.1 ± 2.9	Non-Local Graph Neural Networks
LSC-ARMA	68.4 ± 2.3	Restructuring Graph for Higher Homophily via Adaptive Spectral Clustering	-
FSGNN (3-hop)	78.14±1.25	Improving Graph Neural Networks with Simple Architecture Design
Conn-NSD	65.21±2.04	Sheaf Neural Networks with Connection Laplacians
IIE-GNN	72.13±2.11	Enhancing Intra-class Information Extraction for Heterophilous Graphs: One Neural Architecture Search Approach	-
M2M-GNN	75.20 ± 2.3	Sign is Not a Remedy: Multiset-to-Multiset Message Passing for Learning on Heterophilic Graphs

0 of 61 row(s) selected.