AI Paper Atlas

Research map / Graph learning / GNN architectures

Full-Spectrum Graph Neural Networks: Expressive and Scalable

Xiaohan Wang, Deyu Bo, Longlong Li, Kelin Xia

ICML 2026 · Graph learning › GNN architectures · approach: Weisfeiler-Leman expressivity hierarchies · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add off-diagonal eigenspace interactions (ablation)FSpecGNN(Cheb) · Texasaccuracy: 51.0 -> 55.3Improved
add off-diagonal eigenspace interactions (ablation)FSpecGNN(Cheb) · Squirrelaccuracy: 34.6 -> 39.6Improved
add off-diagonal eigenspace interactions (ablation)FSpecGNN(Cheb) · Roman Empireaccuracy: 53.8 -> 54.4Improved
add in-filter f (ablation)FSpecGNN(Cheb) · Texasaccuracy: 49.5 -> 55.3Improved
add in-filter f (ablation)FSpecGNN(Cheb) · Squirrelaccuracy: 29.9 -> 39.6Improved
add in-filter f (ablation)FSpecGNN(Cheb) · Roman Empireaccuracy: 52.0 -> 54.4Improved
add in-filter f and off-diagonal interactions together (ablation)FSpecGNN(Cheb) · Texasaccuracy: 39.7 -> 55.3Improved
add in-filter f and off-diagonal interactions together (ablation)FSpecGNN(Cheb) · Squirrelaccuracy: 24.5 -> 39.6Improved
add in-filter f and off-diagonal interactions together (ablation)FSpecGNN(Cheb) · Roman Empireaccuracy: 47.1 -> 54.4Improved
add full-spectrum eigenspace interaction to Chebyshev filter (ChebNet to FSpecGNN(Cheb))ChebNet · Squirrelaccuracy: 22.58 -> 39.57Improved
add full-spectrum eigenspace interaction to Chebyshev filter (ChebNet to FSpecGNN(Cheb))ChebNet · Texasaccuracy: 36.76 -> 55.32Improved
add full-spectrum interaction to Bernstein filter (BernNet to FSpecGNN(Bern))BernNet · QuestionsROC-AUC: 65.41 -> 77.11Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy