AI Paper Atlas

Research map / Graph learning / GNN architectures

Hierarchical Multi Scale Graph Neural Networks: Scalable Heterophilous Learning with Oversmoothing and Oversquashing Mitigation

MD SAZZAD Hossen, Avimanyu Sahoo

ICML 2026 · Graph learning › GNN architectures · approach: Oversmoothing and heterophily · 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 adaptive signed affinity instead of fixed adjacency (ablation)HMH · Actoraccuracy (%): 39.8 -> 43.3Improved
add multi-level hierarchy instead of Haar filtering at level 0 only (ablation)HMH · Actoraccuracy (%): 37.5 -> 43.3Improved
use learned localized Haar basis instead of Chebyshev global basis (ablation)HMH · Actoraccuracy (%): 38.1 -> 43.3Improved
add adaptive signed affinity instead of fixed adjacency (ablation)HMH · Chameleon (non-filtered)accuracy (%): 37.0 -> 41.8Improved
add multi-level hierarchy instead of Haar filtering at level 0 only (ablation)HMH · Chameleon (non-filtered)accuracy (%): 34.2 -> 41.8Improved
use learned localized Haar basis instead of Chebyshev global basis (ablation)HMH · Chameleon (non-filtered)accuracy (%): 37.2 -> 41.8Improved
add adaptive signed affinity instead of fixed adjacency (ablation)HMH · Squirrel (non-filtered)accuracy (%): 47.0 -> 50.9Improved
add multi-level hierarchy instead of Haar filtering at level 0 only (ablation)HMH · Squirrel (non-filtered)accuracy (%): 45.4 -> 50.9Improved
use learned localized Haar basis instead of Chebyshev global basis (ablation)HMH · Squirrel (non-filtered)accuracy (%): 48.3 -> 50.9Improved
add adaptive signed affinity instead of fixed adjacency (ablation)HMH · PubMedaccuracy (%): 87.3 -> 91.4Improved
use HMH instead of GCNHMH · Penn94 spokesaccuracy (%): 71.35 -> 82.15Improved
use HMH instead of GIN/ChebNetHMH · Tree-NeighborsMatch depth 10accuracy (%): 50% -> >98%Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 0 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