Diagnosing GNN Failure Modes with Layer-Wise Probe Entropy
arXiv:2610.22595v1 Announce Type: new Abstract: Graph neural networks (GNNs) are typically evaluated only at their final layer, obscuring how task-relevant information evolves during message passing. We introduce a post-hoc diagnostic that freezes a trained GNN, fits an independent linear probe to…
Read the full story at arXiv cs.LG ↗
Timeline · 1 report
- 2026-10-19 04:00 · arXiv cs.LG
Diagnosing GNN Failure Modes with Layer-Wise Probe Entropy