Volume: Vol-4059 EXPLIMED 2025 | Year: 2025

Paper: « paper1 | paper2 | paper3 »

Massive Activations in Graph Neural Networks: Decoding Attention dor Domain-Dependent Interpretability