Graph diffusion models frequently distort substructure count distributions, and the paper shows theoretically and empirically that the GNN backbone's subgraph-counting expressivity controls this behavior.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Do Graph Diffusion Models Accurately Capture and Generate Substructure Distributions?
Graph diffusion models frequently distort substructure count distributions, and the paper shows theoretically and empirically that the GNN backbone's subgraph-counting expressivity controls this behavior.