B-cos GNNs achieve inherent explainability in graph neural networks by using linear aggregation and B-cos transforms to produce exact per-node per-feature contribution decompositions via dynamic linearity.
Title resolution pending
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
-
B-cos GNNs: Faithful Explanations through Dynamic Linearity
B-cos GNNs achieve inherent explainability in graph neural networks by using linear aggregation and B-cos transforms to produce exact per-node per-feature contribution decompositions via dynamic linearity.