HyperAI초신경

Node Classification On Coauthor Physics

평가 지표

Accuracy

평가 결과

이 벤치마크에서 각 모델의 성능 결과

		Paper Title
NCSAGE	98.69 ± 0.26	Clarify Confused Nodes via Separated Learning
NCGCN	98.63 ± 0.24	Clarify Confused Nodes via Separated Learning
GCN	97.46 ± 0.10	Classic GNNs are Strong Baselines: Reassessing GNNs for Node Classification
3ference	97.22%	Inferring from References with Differences for Semi-Supervised Node Classification on Graphs
CoLinkDist	97.05%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
GNNMoE(GAT-like P)	97.05±0.19	Mixture of Experts Meets Decoupled Message Passing: Towards General and Adaptive Node Classification
GNNMoE(GCN-like P)	97.03±0.13	Mixture of Experts Meets Decoupled Message Passing: Towards General and Adaptive Node Classification
LinkDistMLP	96.91%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
Exphormer	96.89±0.09%	Exphormer: Sparse Transformers for Graphs
LinkDist	96.87%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
CoLinkDistMLP	96.87%	Distilling Self-Knowledge From Contrastive Links to Classify Graph Nodes Without Passing Messages
GNNMoE(SAGE-like P)	96.81±0.22	Mixture of Experts Meets Decoupled Message Passing: Towards General and Adaptive Node Classification
GraphMix (GCN)	94.49 ± 0.84	GraphMix: Improved Training of GNNs for Semi-Supervised Learning
DAGNN (Ours)	94	Towards Deeper Graph Neural Networks

0 of 14 row(s) selected.