Salvato in:
Dettagli Bibliografici
Autori principali: Chen, Hongjiang, Zheng, Xin, Jiao, Pengfei, Liu, Huan, Zhao, Zhidong, Wu, Huaming, Xia, Feng, Pan, Shirui
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2605.19822
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916027960393728
author Chen, Hongjiang
Zheng, Xin
Jiao, Pengfei
Liu, Huan
Zhao, Zhidong
Wu, Huaming
Xia, Feng
Pan, Shirui
author_facet Chen, Hongjiang
Zheng, Xin
Jiao, Pengfei
Liu, Huan
Zhao, Zhidong
Wu, Huaming
Xia, Feng
Pan, Shirui
contents Temporal graph neural networks (TGNNs) have gained significant traction for solving real-world temporal graph tasks. However, their interpretability remains limited, as most TGNNs fail to identify which historical interactions most influence a given prediction. Despite promising progress on interpretable TGNNs, existing methods predominantly focus on previously seen historical interactions, which we term stability patterns, while overlooking newly emerging first-time interactions, which we term transition patterns. Both types of patterns are essential for faithful temporal explanations. To address this limitation, we propose ST-TGExplainer, a self-explainable TGNN that disentangles Stability and Transition patterns in temporal graphs for a more faithful Temporal GNN Explainer. Guided by a disentangled information bottleneck objective, ST-TGExplainer learns a compact explanatory subgraph that remains predictive of the event label while explicitly suppressing label-conditioned redundancy between stability and transition patterns. Extensive experiments demonstrate that ST-TGExplainer achieves strong predictive performance and yields more faithful explanations. Code is available at https://github.com/hjchen-hdu/ST-TGExplainer.
format Preprint
id arxiv_https___arxiv_org_abs_2605_19822
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle ST-TGExplainer: Disentangling Stability and Transition Patterns for Temporal GNN Interpretability
Chen, Hongjiang
Zheng, Xin
Jiao, Pengfei
Liu, Huan
Zhao, Zhidong
Wu, Huaming
Xia, Feng
Pan, Shirui
Machine Learning
Artificial Intelligence
Temporal graph neural networks (TGNNs) have gained significant traction for solving real-world temporal graph tasks. However, their interpretability remains limited, as most TGNNs fail to identify which historical interactions most influence a given prediction. Despite promising progress on interpretable TGNNs, existing methods predominantly focus on previously seen historical interactions, which we term stability patterns, while overlooking newly emerging first-time interactions, which we term transition patterns. Both types of patterns are essential for faithful temporal explanations. To address this limitation, we propose ST-TGExplainer, a self-explainable TGNN that disentangles Stability and Transition patterns in temporal graphs for a more faithful Temporal GNN Explainer. Guided by a disentangled information bottleneck objective, ST-TGExplainer learns a compact explanatory subgraph that remains predictive of the event label while explicitly suppressing label-conditioned redundancy between stability and transition patterns. Extensive experiments demonstrate that ST-TGExplainer achieves strong predictive performance and yields more faithful explanations. Code is available at https://github.com/hjchen-hdu/ST-TGExplainer.
title ST-TGExplainer: Disentangling Stability and Transition Patterns for Temporal GNN Interpretability
topic Machine Learning
Artificial Intelligence
url https://arxiv.org/abs/2605.19822