NodePFN pre-trains on synthetic graphs with controllable homophily and causal feature-label models to achieve 71.27 average accuracy on 23 node classification benchmarks without graph-specific training.
Revisiting Graph Neural Networks: All We Have is Low-Pass Filters
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
abstract
Graph neural networks have become one of the most important techniques to solve machine learning problems on graph-structured data. Recent work on vertex classification proposed deep and distributed learning models to achieve high performance and scalability. However, we find that the feature vectors of benchmark datasets are already quite informative for the classification task, and the graph structure only provides a means to denoise the data. In this paper, we develop a theoretical framework based on graph signal processing for analyzing graph neural networks. Our results indicate that graph neural networks only perform low-pass filtering on feature vectors and do not have the non-linear manifold learning property. We further investigate their resilience to feature noise and propose some insights on GCN-based graph neural network design.
verdicts
UNVERDICTED 10representative citing papers
GaRA generates task-specific LoRA weight updates conditioned on graph structures to enable better whole-graph encoding in LLMs for zero-shot graph learning.
Early-exit GNNs for link prediction move the speed-quality Pareto frontier on the HeaRT benchmark by allowing implicit early exiting without auxiliary losses.
HISTOGRAPH applies unified layer-wise attention followed by node-wise attention over historical GNN activations to improve graph classification, especially in deep models.
RED is adapted to graph signals with deep unrolling for parameter estimation, yielding lower MSE than prior graph denoising methods on synthetic and real data.
C3E estimates hidden dimensions and depths for GNNs by treating them as communication channels to reduce over-squashing and improve representation learning.
SUPT assigns prompt features at the subgraph level to enable universal prompt tuning for any GNN pre-training strategy and outperforms fine-tuning in 42 of 45 full-shot and 41 of 45 few-shot graph experiments with average gains of 2.5% and 6.6%.
Robust diffusion operators and hidden-state re-propagation improve PPGNN accuracy to match message-passing GNNs on benchmarks.
GNN generalization depends explicitly on graph structural complexity measured by effective edges, with a new regularization method shown to balance underfitting and overfitting.
TAGR repairs graphs with sparse Gaussian feature-neighborhood edges plus topology-aware residual correction to boost GNN robustness on noisy or incomplete citation networks.
citing papers explorer
-
Learning Posterior Predictive Distributions for Node Classification from Synthetic Graph Priors
NodePFN pre-trains on synthetic graphs with controllable homophily and causal feature-label models to achieve 71.27 average accuracy on 23 node classification benchmarks without graph-specific training.
-
Enhancing LLMs for Graph Tasks via Graph-aware LoRA Generation
GaRA generates task-specific LoRA weight updates conditioned on graph structures to enable better whole-graph encoding in LLMs for zero-shot graph learning.
-
Early-Exit Graph Neural Networks for Link Prediction
Early-exit GNNs for link prediction move the speed-quality Pareto frontier on the HeaRT benchmark by allowing implicit early exiting without auxiliary losses.
-
Learning from Historical Activations in Graph Neural Networks
HISTOGRAPH applies unified layer-wise attention followed by node-wise attention over historical GNN activations to improve graph classification, especially in deep models.
-
Graph Signal Denoising Using Regularization by Denoising and Its Parameter Estimation
RED is adapted to graph signals with deep unrolling for parameter estimation, yielding lower MSE than prior graph denoising methods on synthetic and real data.
-
How Wide and How Deep? Mitigating Over-Squashing of GNNs via Channel Capacity Constrained Estimation
C3E estimates hidden dimensions and depths for GNNs by treating them as communication channels to reduce over-squashing and improve representation learning.
-
Subgraph-level Universal Prompt Tuning
SUPT assigns prompt features at the subgraph level to enable universal prompt tuning for any GNN pre-training strategy and outperforms fine-tuning in 42 of 45 full-shot and 41 of 45 few-shot graph experiments with average gains of 2.5% and 6.6%.
-
Revisiting Pre-Propagation GNNs: Robust Diffusion Operators and Hidden-State Re-Propagation
Robust diffusion operators and hidden-state re-propagation improve PPGNN accuracy to match message-passing GNNs on benchmarks.
-
Rethinking Generalization in Graph Neural Networks: A Structural Complexity Perspective
GNN generalization depends explicitly on graph structural complexity measured by effective edges, with a new regularization method shown to balance underfitting and overfitting.
-
Topology-Aware Gaussian Graph Repair for Robust Graph Neural Networks
TAGR repairs graphs with sparse Gaussian feature-neighborhood edges plus topology-aware residual correction to boost GNN robustness on noisy or incomplete citation networks.