Node Classification On Cora
المقاييس
Accuracy
النتائج
نتائج أداء النماذج المختلفة على هذا المعيار القياسي
جدول المقارنة
| اسم النموذج | Accuracy |
|---|---|
| graph-wavelet-neural-network-1 | 81.6% |
| graph-convolutional-neural-networks-via-1 | 81.9% |
| gresnet-graph-residuals-for-reviving-deep | 85.5% |
| structure-fusion-based-on-graph-convolutional | 83.3% |
| snore-scalable-unsupervised-learning-of | 82.2% |
| is-heterophily-a-real-nightmare-for-graph | 88.95% ± 1.04% |
| tree-decomposed-graph-neural-network | 85.35% ± 0.49% |
| diffwire-inductive-graph-rewiring-via-the | 83.66% |
| every-node-counts-self-ensembling-graph | 83.5% ± 0.4% |
| revisiting-semi-supervised-learning-with | 75.7% |
| understanding-over-squashing-and-bottlenecks-1 | 82.76±0.23% |
| gresnet-graph-residuals-for-reviving-deep | 82.6% |
| cn-motifs-perceptive-graph-neural-networks | 88.20±1.22% |
| fast-graph-representation-learning-with | 82.2% ± 1.5% |
| inferring-from-references-with-differences | 87.78% |
| graph-less-neural-networks-teaching-old-mlps-1 | 80.54± 1.35% |
| transitivity-preserving-graph-representation | 88.74±0.6% |
| convolutional-neural-networks-on-graphs-with | 81.2% |
| deep-graph-infomax | 82.3 ± 0.6% |
| get-rid-of-suspended-animation-problem-deep | 85.1% |
| strong-transitivity-relations-and-graph | - |
| mutual-teaching-for-graph-convolutional | 80.9% |
| tackling-oversmoothing-of-gnns-with-1 | 76.99±1.13% |
| large-scale-learnable-graph-convolutional | 83.3% |
| distilling-self-knowledge-from-contrastive | 88.24% |
| watch-your-step-learning-node-embeddings-via | 67.9% |
| structure-fusion-based-on-graph-convolutional | 83.5% |
| learning-to-make-predictions-on-graphs-with | 78.30% |
| optimization-of-graph-neural-networks-with | 90.16% ± 0.59% |
| graph-optimized-convolutional-networks | 84.8% |
| gresnet-graph-residuals-for-reviving-deep | 83.9% |
| mixhop-higher-order-graph-convolution | 81.9% |
| learning-discrete-structures-for-graph-neural | 84.08 ± 0.4% |
| deep-graph-library-towards-efficient-and | 83.98% ± 0.52% |
| unifying-graph-convolutional-neural-networks-1 | 88.5% ± 1.5% |
| hpgat-high-order-proximity-informed-graph | 83.1% |
| graphnas-graph-neural-architecture-search | 84.2% ± 1.0% |
| graph-u-nets | 84.4% ± 0.6% |
| a-capsule-network-based-model-for-learning | 80.53% |
| gresnet-graph-residuals-for-reviving-deep | 84.3% |
| graph-star-net-for-generalized-multi-task-1 | 82.1% |
| n-gcn-multi-scale-graph-convolution-for-semi | 83.0% |
| adagcn-adaboosting-graph-convolutional | 85.46% ± 0.25% |
| semi-supervised-classification-with-graph | 81.5% |
| deep-graph-contrastive-representation | 83.3% ± 0.4% |
| distilling-self-knowledge-from-contrastive | 87.58% |
| graph-attention-networks | 83.0% ± 0.7% |
| dfnets-spectral-cnns-for-graphs-with-feedback | 86% ± 0.4% |
| splinecnn-fast-geometric-deep-learning-with | 89.48% ± 0.31% |
| certifiable-robustness-and-robust-training | 83% |
| hyperbolic-graph-convolutional-neural | 79.9% |
| graph-adversarial-training-dynamically | 82.6% |
| multi-task-graph-autoencoders | 79.00% |
| measuring-and-relieving-the-over-smoothing | 82.3% |
| graphite-iterative-generative-modeling-of | 82.1% ± 0.06% |
| cleora-a-simple-strong-and-scalable-graph | 86.80% |
| auto-gnn-neural-architecture-search-of-graph | 83.6% ± 0.3% |
| graph-bert-only-attention-is-needed-for | 84.3% |
| predict-then-propagate-graph-neural-networks | 85.29% ± 0.25% |
| diffwire-inductive-graph-rewiring-via-the | 67.96% |
| predict-then-propagate-graph-neural-networks | 85.09% ± 0.25% |
| multi-mask-aggregators-for-graph-neural | 85.80% |
| nodenet-a-graph-regularised-neural-network | 86.80% |
| is-heterophily-a-real-nightmare-for-graph | 89.36% ± 1.26% |
| is-heterophily-a-real-nightmare-for-graph | 88.62% ± 1.22% |
| the-split-matters-flat-minima-methods-for | 88.66 ± 1.38% |
| is-heterophily-a-real-nightmare-for-graph | 88.83% ± 1.49% |
| mitigating-degree-biases-in-message-passing | 87.10±1.53 |
| a-flexible-generative-framework-for-graph | 82.9% |
| distilling-self-knowledge-from-contrastive | 87.54% |
| adaptive-sampling-towards-fast-graph | 87.44% ± 0.0034% |
| distilling-self-knowledge-from-contrastive | 87.89% |