arXiv++ Combinatorics

In many applications, graphs are observed repeatedly, where the task is to observe and track weak localized structural perturbations under noise, such as vertex-coordinate displacements or edge-weight/attribute changes on a fixed vertex set. One-shot graph distances can miss these localized perturbations or fluctuate too strongly to support reliable monitoring, especially in large graphs where signal dilutes with scale. We propose Robust Order-aware Spectral Amplification (ROSA), a distance-amplification operator that integrates metric evaluations along an order-aware edge-removal filtration. We prove that ROSA cannot reduce the included base distance and conditionally increases it whenever a filtered step exposes additional signal; separately, we prove that it can strictly improve an inverse coefficient of variation stability score (IS$^2$), defined as the mean of a noisy graph distance divided by its standard deviation, under an explicit sufficient condition that can be checked empirically. Experiments on synthetic graphs with localized edits under three noise models, varying graph densities, and three algorithmically selected edit sites show that ROSA can often approximately double the IS$^2$ of the base spectral distance. We show where ROSA works and stops working on real-world use cases including mitochondrial networks, fMRI correlation matrices, tissue networks, protein conformer graphs, and retinal vasculature multiplex graphs. The empirical results across graph, operator, and noise settings are consistent with the theoretical amplification conditions and boundary constructions. Evaluated with empirically estimated operative quantities, the bound agrees with the observed gain in direction and approximate scale.

We characterize which local matrix structures saturate Weyl's eigenvalue perturbation bound for graph Laplacians under geometrically constrained vertex displacements. Geometric graphs with heavy-tailed vertex noise arise across sensor networks, biological imaging, and spatial omics, yet tractable predictions for noise-induced spectral error remain limited. We study geometric graphs abstracted from biophysical systems, incorporating clearance, planarity, and identifiability constraints that govern physically realizable embeddings. Within this constrained setting, we identify witness motifs, small subgraphs in maximally noise-sensitive geometric configurations, that dominate weighted-degree and graph Laplacian spectral perturbations under tempered power-law vertex displacements. This motif decomposition reduces global spectral sensitivity to a finite catalog of local extremal structures and identifies configurations that attain Weyl-tight bounds. We then lift these constrained-graph results to general straight-line embedded graphs in arbitrary dimension via local repair operations producing a constrained surrogate graph that preserves sensitivity-relevant structure. To quantify noise-induced spectral variation in both strong-oracle and weak-oracle regimes, we introduce stochastic co-spectrality (SC) and the stochastic spectral separation index (S3I), which characterize when observed spectral distances are noise-driven and when noise parameters are separable. Together, these results provide a principled pathway from local geometric noise to global spectral error in graph Laplacian matrices, enabling estimation of spectral fragility from graph structure without exhaustive eigenvalue computation or restrictive distributional assumptions beyond moment bounds.