Jul 2026
Towards the Explainability of Temporal Graph Networks via Memory Backtracking and Topological Attribution
Experiments show that this method provides faithful explanations and outperforms state-of-the-art baselines on temporal graph datasets, spanning node property prediction, link prediction tasks and graph classification tasks.
Yazheng Liu, Xi Zhang, Si-Hong Xie et al.
· arXiv.org · 0 citations