ST-TGExplainer: Disentangling Stability and Transition Patterns for Temporal GNN Interpretability

Hongjiang Chen, Xin Zheng, Pengfei Jiao, Huan Liu, Zhidong Zhao, Huaming Wu, Feng Xia, Shirui Pan
Proceedings of the 43rd International Conference on Machine Learning, PMLR 306:18568-18588, 2026.

Abstract

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.

Cite this Paper

BibTeX
@InProceedings{pmlr-v306-chen26gx, title = {{ST}-{TGE}xplainer: Disentangling Stability and Transition Patterns for Temporal {GNN} Interpretability}, author = {Chen, Hongjiang and Zheng, Xin and Jiao, Pengfei and Liu, Huan and Zhao, Zhidong and Wu, Huaming and Xia, Feng and Pan, Shirui}, booktitle = {Proceedings of the 43rd International Conference on Machine Learning}, pages = {18568--18588}, year = {2026}, editor = {Zhang, Tong and Dudik, Miroslav and Jaggi, Martin and Agarwal, Alekh and Li, Sharon and Schuurmans, Dale and Zhu, Jerry and Berkenkamp, Felix and Dong, Hanze and Bietti, Alberto}, volume = {306}, series = {Proceedings of Machine Learning Research}, month = {06--11 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v306/main/assets/chen26gx/chen26gx.pdf}, url = {https://proceedings.mlr.press/v306/chen26gx.html}, abstract = {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.} }
Endnote
%0 Conference Paper %T ST-TGExplainer: Disentangling Stability and Transition Patterns for Temporal GNN Interpretability %A Hongjiang Chen %A Xin Zheng %A Pengfei Jiao %A Huan Liu %A Zhidong Zhao %A Huaming Wu %A Feng Xia %A Shirui Pan %B Proceedings of the 43rd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2026 %E Tong Zhang %E Miroslav Dudik %E Martin Jaggi %E Alekh Agarwal %E Sharon Li %E Dale Schuurmans %E Jerry Zhu %E Felix Berkenkamp %E Hanze Dong %E Alberto Bietti %F pmlr-v306-chen26gx %I PMLR %P 18568--18588 %U https://proceedings.mlr.press/v306/chen26gx.html %V 306 %X 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.
APA
Chen, H., Zheng, X., Jiao, P., Liu, H., Zhao, Z., Wu, H., Xia, F. & Pan, S.. (2026). ST-TGExplainer: Disentangling Stability and Transition Patterns for Temporal GNN Interpretability. Proceedings of the 43rd International Conference on Machine Learning, in Proceedings of Machine Learning Research 306:18568-18588 Available from https://proceedings.mlr.press/v306/chen26gx.html.

Related Material