HybSpecNet: A Critical Analysis of Architectural Instability in Hybrid-Domain Spectral GNNs

• URL: http://arxiv.org/abs/2511.16101v1
• Date: Thu, 20 Nov 2025 06:47:44 GMT
• Title: HybSpecNet: A Critical Analysis of Architectural Instability in Hybrid-Domain Spectral GNNs
• Authors: Huseyin Goksu,
• Abstract summary: Spectral Graph Neural Networks offer a principled approach to graph filtering.<n>They face a fundamental "Stability-vs.Adaptivity" trade-off.<n>We propose a hybrid-domain GNN, HybSpecNet, which combines a stable ChebyNet branch with an adaptive KrawtchoukNet branch.
• Score: 0.0
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Spectral Graph Neural Networks offer a principled approach to graph filtering but face a fundamental "Stability-vs-Adaptivity" trade-off. This trade-off is dictated by the choice of spectral domain. Filters in the finite [-1, 1] domain (e.g., ChebyNet) are numerically stable at high polynomial degrees (K) but are static and low-pass, causing them to fail on heterophilic graphs. Conversely, filters in the semi-infinite [0, infty) domain (e.g., KrawtchoukNet) are highly adaptive and achieve SOTA results on heterophily by learning non-low-pass responses. However, as we demonstrate, these adaptive filters can also suffer from numerical instability, leading to catastrophic performance collapse at high K. In this paper, we propose to resolve this trade-off by designing a hybrid-domain GNN, HybSpecNet, which combines a stable `ChebyNet` branch with an adaptive `KrawtchoukNet` branch. We first demonstrate that a "naive" hybrid architecture, which fuses the branches via concatenation, successfully unifies performance at low K, achieving strong results on both homophilic and heterophilic benchmarks. However, we then prove that this naive architecture fails the stability test. Our K-ablation experiments show that this architecture catastrophically collapses at K=25, exactly mirroring the collapse of its unstable `KrawtchoukNet` branch. We identify this critical finding as "Instability Poisoning," where `NaN`/`Inf` gradients from the adaptive branch destroy the training of the model. Finally, we propose and validate an advanced architecture that uses "Late Fusion" to completely isolate the gradient pathways. We demonstrate that this successfully solves the instability problem, remaining perfectly stable up to K=30 while retaining its SOTA performance across all graph types. This work identifies a critical architectural pitfall in hybrid GNN design and provides the robust architectural solution.

Related papers

• Generalizing GNNs with Tokenized Mixture of Experts [75.8310720413187]
  We show that improving stability requires reducing reliance on shift-sensitive features, leaving an irreducible worst-case generalization floor.<n>We propose STEM-GNN, a pretrain-then-finetune framework with a mixture-of-experts encoder for diverse computation paths.<n>Across nine node, link, and graph benchmarks, STEM-GNN achieves a stronger three-way balance, improving robustness to degree/homophily shifts and to feature/edge corruptions while remaining competitive on clean graphs.
  arXiv  Detail & Related papers  (2026-02-09T22:48:30Z)
• LaguerreNet: Advancing a Unified Solution for Heterophily and Over-smoothing with Adaptive Continuous Polynomials [0.0]
  LaguerreNet is a novel GNN filter based on continuous Laguerres Neurals.<n>It is exceptionally robust to over-smoothing, with performance at K=10, an order of magnitude beyond where ChebyNet collapses.
  arXiv  Detail & Related papers  (2025-11-19T10:47:23Z)
• KrawtchoukNet: A Unified GNN Solution for Heterophily and Over-smoothing with Adaptive Bounded Polynomials [0.0]
  Spectral Graph Networks (GNNs) based on filters, such as ChebyNet, suffer from two critical limitations.<n>We propose KrawtchoukNet, a GNN filter based on the discrete Krawtchouks.<n>We show this adaptive nature allows KrawtchoukNet to achieve SOTA performance on challenging benchmarks.
  arXiv  Detail & Related papers  (2025-11-19T10:47:15Z)
• The Final-Stage Bottleneck: A Systematic Dissection of the R-Learner for Network Causal Inference [0.0]
  This paper systematically dissects the R-Learner framework graphs.<n>We provide the first rigorous evidence that the primary driver of performance is the inductive bias of the final-stage CATE estimator.<n>We identify and provide a mechanistic explanation for a subtle, topology-dependent "snuiance bottleneck"
  arXiv  Detail & Related papers  (2025-11-17T06:16:04Z)
• GegenbauerNet: Finding the Optimal Compromise in the GNN Flexibility-Stability Trade-off [0.0]
  Spectral Graph Neural Networks (GNNs) operating in the canonical [-1, 1] domain face a fundamental Flexibility-Stability Trade-off.<n>We propose textbfGegenbauerNet, a novel GNN filter based on the Gegenbauers symmetry.<n>We demonstrate that GegenbauerNet achieves superior performance in the key local filtering regime.
  arXiv  Detail & Related papers  (2025-11-04T19:39:29Z)
• A Signed Graph Approach to Understanding and Mitigating Oversmoothing in GNNs [54.62268052283014]
  We present a unified theoretical perspective based on the framework of signed graphs.<n>We show that many existing strategies implicitly introduce negative edges that alter message-passing to resist oversmoothing.<n>We propose Structural Balanced Propagation (SBP), a plug-and-play method that assigns signed edges based on either labels or feature similarity.
  arXiv  Detail & Related papers  (2025-02-17T03:25:36Z)
• Exact Certification of (Graph) Neural Networks Against Label Poisoning [50.87615167799367]
  We introduce an exact certification method for label flipping in Graph Neural Networks (GNNs)<n>We apply our method to certify a broad range of GNN architectures in node classification tasks.<n>Our work presents the first exact certificate to a poisoning attack ever derived for neural networks.
  arXiv  Detail & Related papers  (2024-11-30T17:05:12Z)
• Graph Structure Learning with Interpretable Bayesian Neural Networks [10.957528713294874]
  We introduce novel iterations with independently interpretable parameters. These parameters influence characteristics of the estimated graph, such as edge sparsity. After unrolling these iterations, prior knowledge over such graph characteristics shape prior distributions. Fast execution and parameter efficiency allow for high-fidelity posterior approximation.
  arXiv  Detail & Related papers  (2024-06-20T23:27:41Z)
• On the Trade-Off between Stability and Representational Capacity in Graph Neural Networks [22.751509906413943]
  We study the stability of EdgeNet: a general GNN framework that unifies more than twenty solutions. By studying the effect of different EdgeNet categories on the stability, we show that GNNs with fewer degrees of freedom in their parameter space, linked to a lower representational capacity, are more stable.
  arXiv  Detail & Related papers  (2023-12-04T22:07:17Z)
• Stability of Aggregation Graph Neural Networks [153.70485149740608]
  We study the stability properties of aggregation graph neural networks (Agg-GNNs) considering perturbations of the underlying graph. We prove that the stability bounds are defined by the properties of the filters in the first layer of the CNN that acts on each node. We also conclude that in Agg-GNNs the selectivity of the mapping operators is tied to the properties of the filters only in the first layer of the CNN stage.
  arXiv  Detail & Related papers  (2022-07-08T03:54:52Z)
• Stability of Neural Networks on Manifolds to Relative Perturbations [118.84154142918214]
  Graph Neural Networks (GNNs) show impressive performance in many practical scenarios. GNNs can scale well on large size graphs, but this is contradicted by the fact that existing stability bounds grow with the number of nodes.
  arXiv  Detail & Related papers  (2021-10-10T04:37:19Z)
• Is Homophily a Necessity for Graph Neural Networks? [50.959340355849896]
  Graph neural networks (GNNs) have shown great prowess in learning representations suitable for numerous graph-based machine learning tasks. GNNs are widely believed to work well due to the homophily assumption ("like attracts like"), and fail to generalize to heterophilous graphs where dissimilar nodes connect. Recent works design new architectures to overcome such heterophily-related limitations, citing poor baseline performance and new architecture improvements on a few heterophilous graph benchmark datasets as evidence for this notion. In our experiments, we empirically find that standard graph convolutional networks (GCNs) can actually achieve better performance than
  arXiv  Detail & Related papers  (2021-06-11T02:44:00Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.