Subgraph Concept Network is a new GNN architecture that distills meaningful concepts at node, subgraph, and graph levels via soft clustering to improve explainability while maintaining competitive accuracy.
This stands in tension with the utilisation loss term, which encourages not all nodes to be assigned to a single cluster
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Subgraph Concept Networks: Concept Levels in Graph Classification
Subgraph Concept Network is a new GNN architecture that distills meaningful concepts at node, subgraph, and graph levels via soft clustering to improve explainability while maintaining competitive accuracy.