Node Classification On Cora

평가 지표

Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

		Paper Title
SSP	90.16% ± 0.59%	Optimization of Graph Neural Networks with Natural Gradient Descent
SplineCNN	89.48% ± 0.31%	SplineCNN: Fast Geometric Deep Learning with Continuous B-Spline Kernels
ACMII-Snowball-3	89.36% ± 1.26%	Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?
ACMII-GCN	88.95% ± 1.04%	Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?
ACM-Snowball-2	88.83% ± 1.49%	Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?
UGT	88.74±0.6%	Transitivity-Preserving Graph Representation Learning for Bridging Local Connectivity and Role-based Similarity
GAT + SWA	88.66 ± 1.38%	The Split Matters: Flat Minima Methods for Improving the Performance of GNNs
ACM-GCN	88.62% ± 1.22%	Is Heterophily A Real Nightmare For Graph Neural Networks To Do Node Classification?
GCN-LPA	88.5% ± 1.5%	Unifying Graph Convolutional Neural Networks and Label Propagation
LinkDist	88.24%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
CNMPGNN	88.20±1.22%	CN-Motifs Perceptive Graph Neural Networks
CoLinkDist	87.89%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
3ference	87.78%	Inferring from References with Differences for Semi-Supervised Node Classification on Graphs
LinkDistMLP	87.58%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
CoLinkDistMLP	87.54%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
AS-GCN	87.44% ± 0.0034%	Adaptive Sampling Towards Fast Graph Representation Learning
CGT	87.10±1.53	Mitigating Degree Biases in Message Passing Mechanism by Utilizing Community Structures
Cleora	86.80%	Cleora: A Simple, Strong and Scalable Graph Embedding Scheme
NodeNet	86.80%	NodeNet: A Graph Regularised Neural Network for Node Classification
DFNet-ATT	86% ± 0.4%	DFNets: Spectral CNNs for Graphs with Feedback-Looped Filters

0 of 72 row(s) selected.

Command Palette

Node Classification On Cora

평가 지표

평가 결과