Node Classification On Squirrel 60 20 20

评估指标

1:1 Accuracy

评测结果

各个模型在此基准测试上的表现结果

		Paper Title	Repository
GNNDLD	77.72±0.84	GNNDLD: Graph Neural Network with Directional Label Distribution	-
ACMII-GCN++	69.98 ± 1.53	Revisiting Heterophily For Graph Neural Networks
ACMII-GCN+	69.81 ± 1.11	Revisiting Heterophily For Graph Neural Networks
ACM-GCN+	69.26 ± 1.11	Revisiting Heterophily For Graph Neural Networks
ACM-GCN++	68.56 ± 1.33	Revisiting Heterophily For Graph Neural Networks
NFGNN	58.9±0.35	Node-oriented Spectral Filtering for Graph Neural Networks
ACM-Snowball-2	55.97 ± 2.03	Revisiting Heterophily For Graph Neural Networks
ACM-Snowball-3	55.73 ± 2.39	Revisiting Heterophily For Graph Neural Networks
ACMII-GCN	54.53 ± 2.09	Revisiting Heterophily For Graph Neural Networks
ACMII-Snowball-2	53.48 ± 0.6	Revisiting Heterophily For Graph Neural Networks
GCN+JK	53.40 ± 1.90	Revisiting Heterophily For Graph Neural Networks
ACMII-Snowball-3	52.31 ± 1.57	Revisiting Heterophily For Graph Neural Networks
GAT+JK	52.28 ± 3.61	Revisiting Heterophily For Graph Neural Networks
BernNet	51.35 ± 0.73	BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation
GPRGNN	49.93 ± 0.53	Adaptive Universal Generalized PageRank Graph Neural Network
Snowball-3	48.25 ± 0.94	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
Snowball-2	47.88 ± 1.23	Break the Ceiling: Stronger Multi-scale Deep Graph Convolutional Networks
SGC-1	47.62 ± 1.27	Simplifying Graph Convolutional Networks
HH-GCN	47.19 ± 1.21	Half-Hop: A graph upsampling approach for slowing down message passing
ACM-SGC-1	46.4 ± 1.13	Revisiting Heterophily For Graph Neural Networks

0 of 37 row(s) selected.