Pith. sign in

REVIEW 4 major objections 5 minor 57 references

Graph Wave Networks

T0 review · 4 major / 5 minor · reviewed 2026-08-07 · deepseek-v4-flash

Challenge this review Re-run · record.json Download PDF Read on arXiv ↗

Pith's one-line read The paper claims that modeling GNN message passing as wave propagation—not heat diffusion—yields a two-step explicit scheme stable at any time step, enabling faster training and stronger accuracy on homophilic and heterophilic graphs.

desk verdict The paper's central stability theorem is false, but the empirical GWN-fa design and careful benchmarking are worth a second look. read the letter →

arxiv 2505.20034 v2 pith:GKZZYNZX submitted 2025-05-26 cs.LG

Juwei Yue , Haikuo Li , Jiawei Sheng , Yihan Guo , Xinghua Zhang , Chuan Zhou , Tingwen Liu , Li Guo This is my paper
classification cs.LG
keywords graphneuralnetworkswaveequationmessagepassingspectralconvolutionPDE-basedlearningstabilityanalysisheterophilyover-smoothing
verification ladder T0 review T1 audit T2 compute T3 formal

The pith

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The reading

This paper argues that message passing in graph neural networks should be modeled as wave propagation rather than heat diffusion, because graph signals are superpositions of eigenvector waves. It derives a graph wave equation from the physics wave equation, discretizes it with the forward Euler method into a two-step explicit recurrence, and claims this recurrence is constantly stable—stable for every positive time step $\tau$. If the claim holds, wave-based GNNs can take far larger time steps than heat-equation models such as GRAND, cutting training cost while keeping or improving accuracy. The paper reports that the resulting networks, GWN-sym and GWN-fa, beat prior methods on homophilic and heterophilic benchmarks, stay accurate as layer count grows, and mitigate over-smoothing.

What carries the argument

The central object is the graph wave equation $\frac{\partial^2 X}{\partial t^2} = L_a X$, in which the graph Laplacian replaces the spatial derivative operator of the physics wave equation, with $L_a = a^2(D - A)$ in discrete form (Proposition 1). The argument is carried by the explicit two-step scheme $X^{(n+1)} = (2I + \tau^2 L_a)X^{(n)} - X^{(n-1)}$, recast as a first-order iteration $U^{(n+1)} = C U^{(n)}$ with iteration matrix $C = \begin{pmatrix} 2I + \tau^2 L_a & -I \\ I & 0 \end{pmatrix}$; Theorem 2 claims that the spectral radius of $C$ never exceeds 1 for $L_a = D^{-1/2}AD^{-1/2}$. That spectral-radius bound is what licenses arbitrarily large time steps and, with them, the paper's efficiency and robustness claims.

What would settle it

Compute the spectral radius of $C = \begin{pmatrix} 2I + \tau^2 D^{-1/2}AD^{-1/2} & -I \\ I & 0 \end{pmatrix}$ for any connected graph (a 10-cycle suffices) at $\tau = 1$: the eigenvalue $\mu = 1$ of $D^{-1/2}AD^{-1/2}$ yields an eigenvalue $(2 + \tau^2 + \sqrt{(2+\tau^2)^2 - 4})/2 > 1$ of $C$, so the iteration grows by that factor per step; running the recurrence from a generic initial condition confirms the norm explosion, which would refute "constantly stable for any $\tau$".

Watch

Extended reading notes

Core claim

The central claim, stated on the paper's own terms, is that a second-order wave equation on the graph is the faithful description of message passing, and that its explicit forward-Euler solution $X^{(n+1)} = (2I + \tau^2 L_a) X^{(n)} - X^{(n-1)}$ is stable for every $\tau \in \mathbb{R}_+$ whenever $L_a = D^{-1/2}AD^{-1/2}$ has eigenvalues in $[-1,1]$ (Theorem 2). The paper connects the wave equation to spectral GNNs by letting the Laplacian play the role of the spatial operator and the wave speed act as either a constant or learnable parameters, yielding two architectures: GWN-sym, built on the symmetric normalized adjacency, and GWN-fa, built on a frequency-adaptive Laplacian with signed attention weights that act as low-pass and high-pass filters. The payoff is that stability no longer constrains the time step, so larger $\tau$ means faster training without an accuracy penalty, and the depth of the network can grow without over-smoothing.

Load-bearing premise

The load-bearing premise is that a two-step recurrence built from any normalized-adjacency eigenvalue between $-1$ and $1$ stays bounded; a graph whose normalized adjacency has an eigenvalue near $+1$ instead produces a growing recurrence at every positive time step, so that premise is what the claim stands on.

Editorial extensions

If this is right

  • If Theorem 2 holds, the time step $\tau$ can be made as large as the solver needs, so the number of iterations—and with it training time—can be cut far below what heat-diffusion models require, which are stable only for tiny steps such as $\tau = 0.005$.
  • Deeper networks become safe: more terminal time $T$ no longer threatens numerical blow-up or over-smoothing, matching the paper's layer-depth experiments on Cora, CiteSeer, and PubMed.
  • Because the equation is defined through the Laplacian, any spectral GNN expressed as a Laplacian filter can be lifted into wave form; the paper demonstrates this by turning GCN into GWN-sym and FAGCN into GWN-fa, each improving on its base model.
  • Signed attention in GWN-fa provides simultaneous low-pass and high-pass behavior, which the paper's heterophilic benchmarks show as large gains over diffusion-based and standard GNN baselines.

Reading between the lines

Editorial extensions of the paper, not claims the author makes directly.

  • Editorial extension: the explicit scheme is the classical leapfrog (Verlet) integrator for second-order dynamics, so the layer stack conserves a discrete energy; tracking that energy during training could serve as an early warning for when a chosen $\tau$ leaves the empirically stable range.
  • Editorial extension: the initial velocity term $\varphi_1(X)$ in Eq. (13) injects momentum; on temporal graphs, feeding the embedding change from the previous snapshot as $\varphi_1$ would yield a natural wave-based model for evolving networks, which the paper does not explore.
  • Editorial extension (testable): the paper's accuracy-versus-$\tau$ curves are flat on large datasets, while smaller heterophilic datasets show a drop at the largest tested step ($\tau = 5$); mapping where that drop begins, relative to the top eigenvalue of $D^{-1/2}AD^{-1/2}$, would pin down the practical operating regime for large-$\tau$ training.
Share X Bluesky LinkedIn Reddit HN

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, and a circularity audit.

Referee Report

4 major / 5 minor

Summary. The paper proposes modeling message passing in GNNs as wave propagation. It derives a graph wave equation, discretizes it with the forward Euler method to obtain an explicit scheme, and claims that the scheme is "constantly stable" for any positive time step (Theorems 2 and 3). Two implementations, GWN-sym and GWN-fa, are introduced and evaluated on node classification benchmarks, with additional experiments on over-smoothing and heterophily.

Significance. If the central stability claim were correct, it would be a significant contribution: explicit schemes for second-order graph PDEs that are unconditionally stable would enable larger time steps and more efficient training than conditionally stable heat-equation methods such as GRAND. The paper also provides a useful connection between wave equations and spectral GNNs, and the experimental section is extensive, including code release and large-scale benchmarks. However, the main theorem is false, and the proof omits a growing eigenmode that is present for every graph with at least one edge. Since the efficiency and robustness advantages are explicitly built on this theorem, the theoretical foundation of the paper is unsupported, even though the empirical results may reflect a useful heuristic.

major comments (4)
  1. [Appendix A.3, Theorem 2] The stability proof for the explicit scheme X^{(n+1)} = (2I + τ²L_a)X^{(n)} − X^{(n−1)} with L_a = D^{−1/2}AD^{−1/2} is incorrect. The proof reduces the second-order recurrence to the characteristic equation λ² − λ′λ + 1 = 0 with λ′ = 2 + τ²μ, μ ∈ [−1,1], and then analyzes only the smaller real root λ₂ for λ′ > 2, together with the complex-root case. For every graph with at least one edge, L_a has an eigenvalue μ = 1 (with eigenvector D^{1/2}1), so λ′ = 2 + τ² > 2 for any τ > 0. In that case the larger real root λ₁ = (2 + τ² + √(τ⁴ + 4τ²))/2 is strictly greater than 1, giving ρ(C) > 1. Thus the scheme is not stable for any positive time step; it diverges exponentially for generic initial data, contradicting the theorem's statement of constant stability.
  2. [Appendix A.5, Theorem 3] The same omission invalidates Theorem 3 for the frequency-adaptive Laplacian. For L_{a,·} = εI ± D^{−1/2}AD^{−1/2} with ε ∈ (0,1), the eigenvalue bound is λ′ ∈ [2 + τ²(ε−1), 2 + τ²(ε+1)]. Since the normalized adjacency has eigenvalue 1, λ′ reaches 2 + τ²(ε+1) > 2 for τ > 0, and the characteristic equation again has a real root λ₁ > 1, so ρ(C) > 1. Additionally, the GWN-fa recurrence actually implemented, Eq. (18), includes the time-dependent attention matrix α^(n) and the ε^(n)X^(0) term; the constant-coefficient stability analysis of Theorem 3 does not apply to this deployed model.
  3. [Proposition 1 / Eq. (11)] The graph wave equation is written as ∂²X/∂t² = a²LX with L = D − A. Since D − A is positive semidefinite, the continuous PDE itself has exponentially growing modes for every positive eigenvalue of L. A standard, well-posed wave equation on a graph requires the Laplacian to have nonpositive spectrum (or the sign in front of L to be negative). This sign choice is not merely a convention: it is exactly what makes both the continuous problem and the discrete scheme unstable, and it is consistent with the failure of Theorem 2.
  4. [Section 4.4 and Tables 7–8] The empirical stability analysis does not support the claimed constant stability. For example, Table 8 reports Texas 48/32/20 accuracy varying from 89.85 (sym-0.2) to 82.75 (sym-5.0), and Cornell 60/20/20 accuracy varying from 89.57 (sym-2.0) to 84.26 (sym-1.0). These are substantial swings across the tested time-step values. The observed behavior is better explained by the bounded terminal time T, early stopping, and the specific hyperparameter tuning than by an unconditional stability property of the scheme.
minor comments (5)
  1. [Appendix A.3 and A.5] The word "Caes" appears twice and should be "Case".
  2. [Figure 7] The figure is referenced to justify assertions about the real roots λ₁ and λ₂, but the text never explains which root is plotted in each panel. The caption should state both roots and explicitly show why λ₁ > 1 for λ′ > 2, which is exactly the case the proof omits.
  3. [Eqs. (13) and (27)] The initial-value notation is inconsistent: Eq. (13) defines φ₀(X) and φ₁(X), while the text after Eq. (27) refers to "φ₁(X) and φ₂(X)".
  4. [Tables 2 and 3] The word "Datesets" in table headers is a typo; also references [34] and [35] are identical and should be merged or corrected.
  5. [Eq. (17)] The objects L_{a,l} and L_{a,h} are called "Laplacians", but they are affine combinations of the identity and the normalized adjacency matrix, not Laplacian matrices in the standard graph-theoretic sense. Consider renaming them "filters" or "operators" to avoid confusion.

Circularity Check

0 steps flagged · score 0.0 of 10

No circularity: the GWN recurrence and stability theorem are derived from the wave equation and evaluated on held-out splits; the proof's mathematical flaw is a correctness issue, not a circular one.

full rationale

The paper's derivation chain is self-contained: the graph wave equation (Eq. 10) is rewritten via the identity L = D - A (Proposition 1), discretized with forward Euler to obtain the explicit scheme (Eq. 14), and then instantiated as GWN-sym (Eq. 16) and GWN-fa (Eq. 18). These are standard algebraic manipulations and finite-difference discretizations; no quantity is defined in terms of the result it is claimed to establish. The stability theorems (Theorem 2 and Theorem 3) are stated as consequences of the companion-matrix spectral analysis in Appendices A.3 and A.5; even if that analysis is mathematically incorrect (the real-root case lambda' > 2 is not treated properly), the stability conclusion is not assumed among the proof's inputs, so the failure is a correctness defect rather than circularity. Empirical results are measured on held-out test splits, and the learnable parameters (attention alpha^(n), epsilon^(n), the projection MLPs) are trained on the supervised loss, not fitted to the benchmark accuracies or to any theorem. Citations are to external baselines and standard graph-signal-processing references, and no load-bearing uniqueness claim is imported from the authors' own prior work. The 'graph wave equation' is a direct graph analogue of the continuum wave equation, so it is a reformulation of a known operator, but it is not a renamed empirical pattern and it does not smuggle in the stability conclusion. Overall, the central derivation and experiments do not reduce to their inputs by construction, and no specific circular reduction can be quoted; the serious concern in this paper is the unsound stability proof, which falls under correctness risk rather than circularity.

Assumptions & free parameters 5 free parameters · 5 assumptions · 0 invented entities

The central stability result rests on the spectral-radius criterion and the eigenvalue interval of the normalized adjacency matrix. The proof misapplies the criterion for real roots; the modeled continuous equation has the wrong sign; and the GWN-fa proof does not cover the actual time-dependent attention update.

free parameters (5)
  • Time step τ = Selected per dataset from {0.2, 0.5, 1.0, 2.0, 5.0}; exact final values not reported in the main text
    Controls the explicit scheme in Eq. (14); the unconditional stability claim is about arbitrary τ, but the experiments tune it.
  • Terminal time T (number of layers) = Uniformly sampled in [1,20] via wandb
    Number of recurrence steps; chosen by hyperparameter search.
  • Frequency scaling ε^{(n)} = Learnable parameter in (0,1)
    Appears in Eq. (17)-(18) of GWN-fa; learned from data.
  • Attention vector g^{(n)} = Learnable vector in R^{2d}
    Defines attention weights α_{ij} in Eq. (18); fitted during training.
  • Hidden dimension d = Chosen from {32, 64, 128, 256}
    Model capacity hyperparameter tuned per dataset.
assumptions (5)
  • domain assumption Graph signals can be treated as a superposition of eigenvector waves with eigenvalue as frequency
    Motivates replacing the heat equation by the wave equation in Remark 1 and Sec. 3.1, but this is a modeling metaphor, not an established fact for arbitrary node features.
  • domain assumption Message passing can be modeled as a continuous PDE in time over node embeddings and solved by finite differencing
    Sec. 3.3 inherits this from GRAND-style PDE GNNs; the truncation error is discarded without control.
  • domain assumption The graph Laplacian operator Δ is represented by L = D − A and yields ∂²X/∂t² = a²LX
    Proposition 1; the sign convention makes the spatial operator positive semidefinite, causing growing exponential modes instead of waves.
  • standard math Spectral radius criterion ρ(C) ≤ 1 is sufficient for stability of the iteration
    Theorem 1 in Sec. 3.5; for non-normal matrices spectral radius is necessary but not sufficient, and C is non-symmetric in U-space.
  • standard math Eigenvalues of D^{−1/2}AD^{−1/2} lie in [−1,1]
    Used in Theorems 2 and 3; true by spectral graph theory, but the conclusion drawn from it is false.

how reviews work

0 comments
Cite this review

Pith. "Pith review of Graph Wave Networks." pith.science (2026). https://pith.science/paper/GKZZYNZX

@misc{pith2026250520034,
  author       = {Pith},
  title        = {Pith review of: Graph Wave Networks},
  year         = {2026},
  howpublished = {\url{https://pith.science/paper/GKZZYNZX}},
  note         = {Machine review of arXiv:2505.20034}
}
read the original abstract

Dynamics modeling has been introduced as a novel paradigm in message passing (MP) of graph neural networks (GNNs). Existing methods consider MP between nodes as a heat diffusion process, and leverage heat equation to model the temporal evolution of nodes in the embedding space. However, heat equation can hardly depict the wave nature of graph signals in graph signal processing. Besides, heat equation is essentially a partial differential equation (PDE) involving a first partial derivative of time, whose numerical solution usually has low stability, and leads to inefficient model training. In this paper, we would like to depict more wave details in MP, since graph signals are essentially wave signals that can be seen as a superposition of a series of waves in the form of eigenvector. This motivates us to consider MP as a wave propagation process to capture the temporal evolution of wave signals in the space. Based on wave equation in physics, we innovatively develop a graph wave equation to leverage the wave propagation on graphs. In details, we demonstrate that the graph wave equation can be connected to traditional spectral GNNs, facilitating the design of graph wave networks based on various Laplacians and enhancing the performance of the spectral GNNs. Besides, the graph wave equation is particularly a PDE involving a second partial derivative of time, which has stronger stability on graphs than the heat equation that involves a first partial derivative of time. Additionally, we theoretically prove that the numerical solution derived from the graph wave equation are constantly stable, enabling to significantly enhance model efficiency while ensuring its performance. Extensive experiments show that GWNs achieve SOTA and efficient performance on benchmark datasets, and exhibit outstanding performance in addressing challenging graph problems, such as over-smoothing and heterophily.

Figures

Figures reproduced from arXiv: 2505.20034 by the authors.

Figure 1
Figure 1. Top: the partial spectrum on the real-world dataset [PITH_FULL_IMAGE:figures/full_fig_p002_1.png] view at source ↗
Figure 2
Figure 2. Visualization of the attention matrix 𝜶 of GWN-fa. indicates that GWN outperforms heat equation based methods sush as GRAND on 5 datasets and achieves suboptimal performance on Photo. Since both GWN-sym and GWN-fa can be treated as low-pass filters, their performances are closely comparable. Then, we examine the generality of GWN on heterophilic graphs [PITH_FULL_IMAGE:figures/full_fig_p006_2.png] view at source ↗
Figure 3
Figure 3. Performances of methods of each layer on citation networks. [PITH_FULL_IMAGE:figures/full_fig_p007_3.png] view at source ↗
Figures from the paper (5 more)
Figure 4
Figure 4. Figure 4: Stability analysis of GWN. Red dashed line indicates the SOTA models. [PITH_FULL_IMAGE:figures/full_fig_p007_4.png]
Figure 5
Figure 5. Figure 5: Visualization of waveform in wave propagation of [PITH_FULL_IMAGE:figures/full_fig_p008_5.png]
Figure 6
Figure 6. Figure 6: Visualization of time and node view in wave propa [PITH_FULL_IMAGE:figures/full_fig_p008_6.png]
Figure 7
Figure 7. Figure 7: The function curve of real roots 𝜆1 (Left) and 𝜆2 (Right). And if Δ = 𝜆 ′2 − 4 ≥ 0, that is |𝜆 ′ | ≥ 2, then the equation has a pair of real roots 𝜆1 = 𝜆 ′+ √ 𝜆 ′2−4 2 and 𝜆2 = 𝜆 ′− √ 𝜆 ′2−4 2 . We next discuss whether the |𝜆| ≤ 1 satisfies for different 𝜏, considering…
Figure 8
Figure 8. Figure 8: Accuracy and runtime of the methods: "sym" and "fa" denote two variants of GWN; "l", "nl", and "nl-rw" denote three [PITH_FULL_IMAGE:figures/full_fig_p014_8.png]

Discussion (0). Sign in to comment.

Reference graph

Works this paper leans on

57 extracted references · 37 canonical work pages

  1. [1]

    Sami Abu-El-Haija, Bryan Perozzi, Amol Kapoor, Nazanin Alipourfard, Kristina Lerman, Hrayr Harutyunyan, Greg Ver Steeg, and Aram Galstyan. 2019. MixHop: Higher-Order Graph Convolutional Architectures via Sparsified Neighborhood Mixing. InProceedings of the 36th International Conference on Machine Learning, ICML 2019, 9-15 June 2019, Long Beach, California...

    work page 2019

  2. [2]

    Filippo Maria Bianchi, Daniele Grattarola, Lorenzo Livi, and Cesare Alippi. 2022. Graph Neural Networks With Convolutional ARMA Filters.IEEE Trans. Pattern Anal. Mach. Intell.44, 7 (2022), 3496–3507. doi:10.1109/TPAMI.2021.3054830

    arXiv 2022

  3. [3]

    Deyu Bo, Xiao Wang, Chuan Shi, and Huawei Shen. 2021. Beyond Low-frequency Information in Graph Convolutional Networks. InThirty-Fifth AAAI Conference on Artificial Intelligence, AAAI 2021, Thirty-Third Conference on Innovative Appli- cations of Artificial Intelligence, IAAI 2021, The Eleventh Symposium on Educational Advances in Artificial Intelligence, ...

    work page 2021

  4. [4]

    Bronstein

    Cristian Bodnar, Francesco Di Giovanni, Benjamin Paul Chamberlain, Pietro Lió, and Michael M. Bronstein. 2022. Neural Sheaf Diffusion: A Topological Perspective on Heterophily and Oversmoothing in GNNs. InNeurIPS

    work page 2022

  5. [5]

    Joan Bruna, Wojciech Zaremba, Arthur Szlam, and Yann LeCun. 2014. Spectral Networks and Locally Connected Networks on Graphs. In2nd International Conference on Learning Representations, ICLR 2014, Banff, AB, Canada, April 14- 16, 2014, Conference Track Proceedings, Yoshua Bengio and Yann LeCun (Eds.). http://arxiv.org/abs/1312.6203

    arXiv 2014

  6. [6]

    Gorinova, Michael M

    Ben Chamberlain, James Rowbottom, Maria I. Gorinova, Michael M. Bronstein, Stefan Webb, and Emanuele Rossi. 2021. GRAND: Graph Neural Diffusion. In Proceedings of the 38th International Conference on Machine Learning, ICML 2021, 18-24 July 2021, Virtual Event (Proceedings of Machine Learning Research, Vol. 139), Marina Meila and Tong Zhang (Eds.). PMLR, 1...

    work page 2021

  7. [7]

    Ming Chen, Zhewei Wei, Zengfeng Huang, Bolin Ding, and Yaliang Li. 2020. Simple and Deep Graph Convolutional Networks. InProceedings of the 37th International Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event. PMLR

    work page 2020

  8. [8]

    Tian Qi Chen, Yulia Rubanova, Jesse Bettencourt, and David Duvenaud. 2018. Neural Ordinary Differential Equations. InAdvances in Neural Information Pro- cessing Systems 31: Annual Conference on Neural Information Processing Systems 2018, NeurIPS 2018, December 3-8, 2018, Montréal, Canada, Samy Bengio, Hanna M. Wallach, Hugo Larochelle, Kristen Grauman, Ni...

    work page 2018

Show all 57 references
  1. [9]

    Eli Chien, Jianhao Peng, Pan Li, and Olgica Milenkovic. 2021. Adaptive Universal Generalized PageRank Graph Neural Network. In9th International Conference on Learning Representations, ICLR 2021, Virtual Event, Austria, May 3-7, 2021. OpenReview.net

    2021

  2. [10]

    Michaël Defferrard, Xavier Bresson, and Pierre Vandergheynst. 2016. Convo- lutional Neural Networks on Graphs with Fast Localized Spectral Filtering. In Advances in Neural Information Processing Systems 29: Annual Conference on Neu- ral Information Processing Systems 2016, Dec...

    2016

  3. [11]

    Moshe Eliasof, Eldad Haber, and Eran Treister. 2021. PDE-GCN: Novel Architec- tures for Graph Neural Networks Motivated by Partial Differential Equations. InAdvances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021, N...

    2021

  4. [12]

    Matthias Fey and Jan Eric Lenssen. 2019. Fast Graph Representation Learning with PyTorch Geometric. InICLR Workshop on Representation Learning on Graphs and Manifolds

    2019

  5. [13]

    Jamasb, Jyothish Soman, Cristian Regep, Gertrude Liu, Jeremy B

    Thomas Gaudelet, Ben Day, Arian R. Jamasb, Jyothish Soman, Cristian Regep, Gertrude Liu, Jeremy B. R. Hayter, Richard Vickers, Charles Roberts, Jian Tang, David Roblin, Tom L. Blundell, Michael M. Bronstein, and Jake P. Taylor-King

  6. [14]

    Haoyu Geng, Chao Chen, Yixuan He, Gang Zeng, Zhaobing Han, Hua Chai, and Junchi Yan. 2023. Pyramid Graph Neural Network: A Graph Sampling and Filtering Approach for Multi-scale Disentangled Representations. InProceedings of the 29th ACM SIGKDD Conference on Knowledge Discovery...

    2023

  7. [15]

    Hamilton, Zhitao Ying, and Jure Leskovec

    William L. Hamilton, Zhitao Ying, and Jure Leskovec. 2017. Inductive Represen- tation Learning on Large Graphs. InAdvances in Neural Information Processing Systems 30: Annual Conference on Neural Information Processing Systems 2017, December 4-9, 2017, Long Beach, CA, USA

    2017

  8. [16]

    Mingguo He, Zhewei Wei, Zengfeng Huang, and Hongteng Xu. 2021. BernNet: Learning Arbitrary Graph Spectral Filters via Bernstein Approximation. InAd- vances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021, NeurIPS 202...

    2021

  9. [17]

    Mingguo He, Zhewei Wei, and Ji-Rong Wen. 2022. Convolutional Neural Net- works on Graphs with Chebyshev Approximation, Revisited. InAdvances in Neural Information Processing Systems 35: Annual Conference on Neural Informa- tion Processing Systems 2022, NeurIPS 2022, New Orlean...

    2022

  10. [18]

    Weihua Hu, Matthias Fey, Marinka Zitnik, Yuxiao Dong, Hongyu Ren, Bowen Liu, Michele Catasta, and Jure Leskovec. 2020. Open Graph Benchmark: Datasets for Machine Learning on Graphs. InAdvances in Neural Information Processing Sys- tems 33: Annual Conference on Neural Informati...

    2020

  11. [19]

    Kingma and Jimmy Ba

    Diederik P. Kingma and Jimmy Ba. 2015. Adam: A Method for Stochastic Opti- mization. In3rd International Conference on Learning Representations, ICLR 2015, San Diego, CA, USA, May 7-9, 2015, Conference Track Proceedings

    2015

  12. [20]

    Kipf and Max Welling

    Thomas N. Kipf and Max Welling. 2017. Semi-Supervised Classification with Graph Convolutional Networks. In5th International Conference on Learning Representations, ICLR 2017, Toulon, France, April 24-26, 2017, Conference Track Proceedings. OpenReview.net

    2017

  13. [21]

    Johannes Klicpera, Stefan Weißenberger, and Stephan Günnemann. 2019. Dif- fusion Improves Graph Learning. InAdvances in Neural Information Processing Systems 32: Annual Conference on Neural Information Processing Systems 2019, NeurIPS 2019, December 8-14, 2019, Vancouver, BC, ...

    2019

  14. [22]

    Thabet, and Bernard Ghanem

    Guohao Li, Matthias Müller, Ali K. Thabet, and Bernard Ghanem. 2019. Deep- GCNs: Can GCNs Go As Deep As CNNs?. In2019 IEEE/CVF International Confer- ence on Computer Vision, ICCV 2019, Seoul, Korea (South), October 27 - November 2, 2019. IEEE, 9266–9275. doi:10.1109/ICCV.2019.00936

    2019

  15. [23]

    Xiang Li, Renyu Zhu, Yao Cheng, Caihua Shan, Siqiang Luo, Dongsheng Li, and Weining Qian. 2022. Finding Global Homophily in Graph Neural Networks When Meeting Heterophily. InInternational Conference on Machine Learning, ICML 2022, 17-23 July 2022, Baltimore, Maryland, USA (Pro...

    2022

  16. [24]

    Yibo Li, Xiao Wang, Hongrui Liu, and Chuan Shi. 2024. A Generalized Neural Diffusion Framework on Graphs. InThirty-Eighth AAAI Conference on Artificial Intelligence, AAAI 2024, Thirty-Sixth Conference on Innovative Applications of Artificial Intelligence, IAAI 2024, Fourteenth...

    2024 doi

  17. [25]

    Kovachki, Kamyar Azizzadenesheli, Burigede Liu, An- drew M

    Zongyi Li, Nikola B. Kovachki, Kamyar Azizzadenesheli, Burigede Liu, An- drew M. Stuart, Kaushik Bhattacharya, and Anima Anandkumar. 2020. Mul- tipole Graph Neural Operator for Parametric Partial Differential Equations. In Advances in Neural Information Processing Systems 33: ...

    2020

  18. [26]

    Derek Lim, Felix Hohne, Xiuyu Li, Sijia Linda Huang, Vaishnavi Gupta, Omkar Bhalerao, and Ser-Nam Lim. 2021. Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methods. InAdvances in Neural Information Processing Systems 34: Annual Conference on N...

    2021

  19. [27]

    Jianghao Lin, Weiwen Liu, Xinyi Dai, Weinan Zhang, Shuai Li, Ruiming Tang, Xiuqiang He, Jianye Hao, and Yong Yu. 2021. A Graph-Enhanced Click Model for Web Search. InSIGIR ’21: The 44th International ACM SIGIR Conference on Research and Development in Information Retrieval, Vi...

    2021

  20. [28]

    Meng Liu, Zhengyang Wang, and Shuiwang Ji. 2022. Non-Local Graph Neural Networks.IEEE Trans. Pattern Anal. Mach. Intell.44, 12 (2022), 10270–10276. doi:10.1109/TPAMI.2021.3134200

    2022

  21. [29]

    Sitao Luan, Chenqing Hua, Qincheng Lu, Jiaqi Zhu, Mingde Zhao, Shuyuan Zhang, Xiao-Wen Chang, and Doina Precup. 2022. Revisiting Heterophily For Graph Neural Networks. InNeurIPS. WWW ’25, April 28-May 2, 2025, Sydney, NSW, Australia Juwei Yue et al

    2022

  22. [30]

    Tuan Nguyen, Hirotada Honda, Takashi Sano, Vinh Nguyen, Shugo Nakamura, and Tan Minh Nguyen. 2024. From Coupled Oscillators to Graph Neural Net- works: Reducing Over-smoothing via a Kuramoto Model-based Approach. In International Conference on Artificial Intelligence and Stati...

    2024

  23. [31]

    Hongbin Pei, Bingzhe Wei, Kevin Chen-Chuan Chang, Yu Lei, and Bo Yang. 2020. Geom-GCN: Geometric Graph Convolutional Networks. In8th International Conference on Learning Representations, ICLR 2020, Addis Ababa, Ethiopia, April 26-30, 2020. OpenReview.net

    2020

  24. [32]

    Michael Poli, Stefano Massaroli, Junyoung Park, Atsushi Yamashita, Hajime Asama, and Jinkyoo Park. 2019. Graph Neural Ordinary Differential Equations. CoRRabs/1911.07532 (2019). arXiv:1911.07532 http://arxiv.org/abs/1911.07532

    2019 arXiv

  25. [33]

    Benedek Rozemberczki, Carl Allen, and Rik Sarkar. 2019. Multi-scale Attributed Node Embedding.CoRRabs/1909.13021 (2019). arXiv:1909.13021

    2019 arXiv

  26. [35]

    Benedek Rozemberczki and Rik Sarkar. 2020. Characteristic Functions on Graphs: Birds of a Feather, from Statistical Descriptors to Parametric Models. InCIKM ’20: The 29th ACM International Conference on Information and Knowledge Man- agement, Virtual Event, Ireland, October 19...

    2020 doi

  27. [36]

    Konstantin Rusch, Ben Chamberlain, James Rowbottom, Siddhartha Mishra, and Michael M

    T. Konstantin Rusch, Ben Chamberlain, James Rowbottom, Siddhartha Mishra, and Michael M. Bronstein. 2022. Graph-Coupled Oscillator Networks. InInter- national Conference on Machine Learning, ICML 2022, 17-23 July 2022, Baltimore, Maryland, USA (Proceedings of Machine Learning ...

    2022

  28. [37]

    Oleksandr Shchur, Maximilian Mumme, Aleksandar Bojchevski, and Stephan Günnemann. 2018. Pitfalls of Graph Neural Network Evaluation.CoRR abs/1811.05868 (2018). arXiv:1811.05868

    2018 arXiv

  29. [38]

    Battaglia, and Jean-Roch Vlimant

    Jonathan Shlomi, Peter W. Battaglia, and Jean-Roch Vlimant. 2021. Graph neural networks in particle physics.Mach. Learn. Sci. Technol.2, 2 (2021), 21001. doi:10. 1088/2632-2153/ABBF9A

    2021

  30. [39]

    Shuman, Sunil K

    David I. Shuman, Sunil K. Narang, Pascal Frossard, Antonio Ortega, and Pierre Vandergheynst. 2013. The Emerging Field of Signal Processing on Graphs: Extend- ing High-Dimensional Data Analysis to Networks and Other Irregular Domains. IEEE Signal Process. Mag.30, 3 (2013), 83–9...

    2013

  31. [40]

    Yunchong Song, Chenghu Zhou, Xinbing Wang, and Zhouhan Lin. 2023. Ordered GNN: Ordering Message Passing to Deal with Heterophily and Over-smoothing. InThe Eleventh International Conference on Learning Representations, ICLR 2023, Kigali, Rwanda, May 1-5, 2023. OpenReview.net. h...

    2023

  32. [41]

    Susheel Suresh, Vinith Budde, Jennifer Neville, Pan Li, and Jianzhu Ma. 2021. Breaking the Limit of Graph Neural Networks by Improving the Assortativity of Graphs with Local Mixing Patterns. InKDD ’21: The 27th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Virt...

    2021

  33. [42]

    Bertozzi, Stanley J

    Matthew Thorpe, Tan Minh Nguyen, Hedi Xia, Thomas Strohmer, Andrea L. Bertozzi, Stanley J. Osher, and Bao Wang. 2022. GRAND++: Graph Neural Diffusion with A Source Term. InThe Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 20...

    2022

  34. [43]

    Petar Velickovic, Guillem Cucurull, Arantxa Casanova, Adriana Romero, Pietro Liò, and Yoshua Bengio. 2018. Graph Attention Networks. In6th International Conference on Learning Representations, ICLR 2018, Vancouver, BC, Canada, April 30 - May 3, 2018, Conference Track Proceedin...

    2018

  35. [44]

    Xiyuan Wang and Muhan Zhang. 2022. How Powerful are Spectral Graph Neural Networks. InInternational Conference on Machine Learning, ICML 2022, 17-23 July 2022, Baltimore, Maryland, USA (Proceedings of Machine Learning Research, Vol. 162), Kamalika Chaudhuri, Stefanie Jegelka, ...

    2022

  36. [45]

    Yifei Wang, Yisen Wang, Jiansheng Yang, and Zhouchen Lin. 2021. Dissecting the Diffusion Process in Linear Graph Convolutional Networks. InAdvances in Neural Information Processing Systems 34: Annual Conference on Neural Information Processing Systems 2021, NeurIPS 2021, Decem...

    2021

  37. [46]

    Souza Jr., Tianyi Zhang, Christopher Fifty, Tao Yu, and Kilian Q

    Felix Wu, Amauri H. Souza Jr., Tianyi Zhang, Christopher Fifty, Tao Yu, and Kilian Q. Weinberger. 2019. Simplifying Graph Convolutional Networks. In Proceedings of the 36th International Conference on Machine Learning, ICML 2019, 9-15 June 2019, Long Beach, California, USA. PMLR

    2019

  38. [47]

    Shiwen Wu, Fei Sun, Wentao Zhang, Xu Xie, and Bin Cui. 2023. Graph Neural Networks in Recommender Systems: A Survey.ACM Comput. Surv.55, 5 (2023), 97:1–97:37. doi:10.1145/3535101

    2023 doi

  39. [48]

    Zonghan Wu, Shirui Pan, Fengwen Chen, Guodong Long, Chengqi Zhang, and Philip S. Yu. 2021. A Comprehensive Survey on Graph Neural Networks.IEEE Trans. Neural Networks Learn. Syst.32, 1 (2021), 4–24. doi:10.1109/TNNLS.2020. 2978386

    2021 doi

  40. [49]

    Louis-Pascal A. C. Xhonneux, Meng Qu, and Jian Tang. 2020. Continuous Graph Neural Networks. InProceedings of the 37th International Conference on Machine Learning, ICML 2020, 13-18 July 2020, Virtual Event (Proceedings of Machine Learning Research, Vol. 119). PMLR, 10432–1044...

    2020

  41. [50]

    Keyulu Xu, Chengtao Li, Yonglong Tian, Tomohiro Sonobe, Ken-ichi Kawarabayashi, and Stefanie Jegelka. 2018. Representation Learning on Graphs with Jumping Knowledge Networks. InProceedings of the 35th International Con- ference on Machine Learning, ICML 2018, Stockholmsmässan,...

    2018

  42. [51]

    Yujun Yan, Milad Hashemi, Kevin Swersky, Yaoqing Yang, and Danai Koutra

  43. [52]

    Cohen, and Ruslan Salakhutdinov

    Zhilin Yang, William W. Cohen, and Ruslan Salakhutdinov. 2016. Revisiting Semi-Supervised Learning with Graph Embeddings. InProceedings of the 33nd International Conference on Machine Learning, ICML 2016, New York City, NY, USA, June 19-24, 2016. JMLR.org

    2016

  44. [53]

    Wentao Zhang, Zeang Sheng, Ziqi Yin, Yuezihan Jiang, Yikuan Xia, Jun Gao, Zhi Yang, and Bin Cui. 2022. Model Degradation Hinders Deep Graph Neural Networks. InKDD ’22: The 28th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, Washington, DC, USA, August 14 - 18, 2...

    2022

  45. [54]

    Jialin Zhao, Yuxiao Dong, Ming Ding, Evgeny Kharlamov, and Jie Tang. 2021. Adaptive Diffusion in Graph Neural Networks. InAdvances in Neural Informa- tion Processing Systems 34: Annual Conference on Neural Information Process- ing Systems 2021, NeurIPS 2021, December 6-14, 202...

    2021

  46. [55]

    Xuebin Zheng, Bingxin Zhou, Junbin Gao, Yuguang Wang, Pietro Lió, Ming Li, and Guido Montúfar. 2021. How Framelets Enhance Graph Neural Networks. InProceedings of the 38th International Conference on Machine Learning, ICML 2021, 18-24 July 2021, Virtual Event (Proceedings of M...

    2021

  47. [56]

    sym-1.0 (1.7x)

    Jiong Zhu, Yujun Yan, Lingxiao Zhao, Mark Heimann, Leman Akoglu, and Danai Koutra. 2020. Beyond Homophily in Graph Neural Networks: Current Limitations and Effective Designs. InAdvances in Neural Information Processing Systems 33: Annual Conference on Neural Information Proces...

    2020

  48. [2021]

    Briefings Bioinform.22, 6 (2021)

    Utilizing graph machine learning within drug discovery and development. Briefings Bioinform.22, 6 (2021). doi:10.1093/BIB/BBAB159

    2021 doi

  49. [2022]

    InIEEE International Conference on Data Mining, ICDM 2022, Orlando, FL, USA, November 28 - Dec

    Two Sides of the Same Coin: Heterophily and Oversmoothing in Graph Convolutional Neural Networks. InIEEE International Conference on Data Mining, ICDM 2022, Orlando, FL, USA, November 28 - Dec. 1, 2022. IEEE. doi:10.1109/ ICDM54844.2022.00169

    2022

Pith tools

Reviewed August 7, 2026 · model on record in the stance chip above.