Node Classification On Coauthor Physics
评估指标
Accuracy
评测结果
各个模型在此基准测试上的表现结果
比较表格
| 模型名称 | Accuracy |
|---|---|
| mixture-of-experts-meets-decoupled-message | 97.03±0.13 |
| distilling-self-knowledge-from-contrastive | 96.87% |
| distilling-self-knowledge-from-contrastive | 97.05% |
| exphormer-sparse-transformers-for-graphs | 96.89±0.09% |
| mixture-of-experts-meets-decoupled-message | 96.81±0.22 |
| towards-deeper-graph-neural-networks | 94 |
| classic-gnns-are-strong-baselines-reassessing | 97.46 ± 0.10 |
| inferring-from-references-with-differences | 97.22% |
| graphmix-regularized-training-of-graph-neural | 94.49 ± 0.84 |
| distilling-self-knowledge-from-contrastive | 96.87% |
| distilling-self-knowledge-from-contrastive | 96.91% |
| mixture-of-experts-meets-decoupled-message | 97.05±0.19 |
| clarify-confused-nodes-through-separated | 98.69 ± 0.26 |
| clarify-confused-nodes-through-separated | 98.63 ± 0.24 |