Back to feed
arXiv cs.LG
arXiv cs.LG
7/10/2026
Towards the Explainability of Temporal Graph Networks via Memory Backtracking and Topological Attribution

Towards the Explainability of Temporal Graph Networks via Memory Backtracking and Topological Attribution

Short summary

MemExplainer is a method for interpreting Temporal Graph Network predictions by tracking how historical events shape node memory vectors through topology attribution and memory backtracking trees. Validated across nine datasets spanning node prediction, link prediction, and graph classification, the approach outperforms existing baselines in explanation faithfulness. Open-source code is available on GitHub.

  • •Explains TGN predictions by tracing historical event influence on node memory vectors
  • •Combines topology attribution trees and memory backtracking with Layer-wise Relevance Propagation
  • •Validated on 9 datasets; outperforms baselines; code publicly available

Generated with AI, which can make mistakes.

#research-breakthrough#ai-tools
Read full article at arXiv cs.LG

Is this a good recommendation for you?

Comments

Failed to load comments. Please try again.

Explore more

CIExplainer++: Generating Causal and Interpretable Explanations for Graph Neural Networks
arXiv cs.LG
From Graphs to Gradients: Physics-Inspired Structural Attribution for Cyber-Physical IoT Systems and Beyond
arXiv CS.AI
Causal-Audit: Explicit and Auditable Graph-based Reasoning via Target-Aware Causal Chain Construction
arXiv CS.AI
The user wants me to rewrite a news headline for a mobile feed based on the original title and summary.
arXiv CS.AI