Modelpedia
Work in progress
About
Findings
Models
Concepts
Methods
Datasets
Sources
Light
Dark
On the Hölder Stability of Multiset and Graph Neural Networks
2025-01-22
· ICLR 2025 Oral ·
anchor
Findings
IC-229
GIN, GCN, and GAT achieve only random-chance accuracy on WL-separable ε-tree graph pairs, exposing a gap between theoretical expressivity and practical separation