Research map / Graph learning / Knowledge and temporal graphs
TempME: Towards the Explainability of Temporal Graph Neural Networks via Motif Discovery
Jialin Chen, Zhitao Ying
Read the paperCheck your idea against 137,217 papers
AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.
Measured effects reported in the paper
| What was changed | Where | Measured | Effect |
|---|---|---|---|
| replace GINE convolution with GCN in motif encoder (ablation) | TempME · Wikipedia, TGAT | ACC-AUC: 85.81 -> 83.26 | Hurt |
| replace GINE convolution with GAT (ablation) | TempME · Wikipedia, TGAT | ACC-AUC: 85.81 -> 84.37 | Hurt |
| replace mean-pooling with adaptive pooling (ablation) | TempME · Wikipedia, TGAT | ACC-AUC: 85.81 -> 85.24 | Hurt |
| remove event anonymization (ablation) | TempME · Wikipedia, TGAT | ACC-AUC: 85.81 -> 81.47 | Hurt |
| remove time encoding (ablation) | TempME · Wikipedia, TGAT | ACC-AUC: 85.81 -> 79.42 | Hurt |
| use uniform prior instead of empirical prior (ablation) | TempME · Wikipedia, GraphMixer | ACC-AUC: 90.15 -> 87.46 | Hurt |
| use uniform prior instead of empirical prior (ablation) | TempME · Wikipedia, TGAT | ACC-AUC: 85.81 -> 85.60 | Hurt |
| add motif embedding to TGAT | TGAT · USLegis | average precision: 72.35 -> 95.31 | Improved |
| add motif embedding to TGN | TGN · Enron | average precision: 76.40 -> 75.62 | Hurt |
| add motif embedding to TGN | TGN · Can.Parl. | average precision: 64.23 -> 79.46 | Improved |
| add motif embedding to GraphMixer | GraphMixer · UCI | average precision: 89.13 -> 90.11 | Improved |
| TempME vs TGNNExplainer when explaining TGAT on Enron | TempME · Enron, TGAT | ACC-AUC: 82.02 -> 81.85 | No clear effect |
Later papers that compared with this method
2 later accepted papers report TempME as a baseline in their results tables; 2 of them report a better number on most of the results they share with it.
- Training-free Counterfactual Explanation for Temporal Graph Model Inference (ICLR 2026 · better on 5 of 5 shared results)
- ST-TGExplainer: Disentangling Stability and Transition Patterns for Temporal GNN Interpretability (ICML 2026 · better on 4 of 4 shared results)
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- CoDy: Counterfactual Explainers for Dynamic Graphs (ICML 2025 · similarity 0.89)
- TAMI: Taming Heterogeneity in Temporal Interactions for Temporal Graph Link Prediction (NeurIPS 2025 · similarity 0.87)
- State Space Models on Temporal Graphs: A First-Principles Study (NeurIPS 2024 · similarity 0.86)
- Improving Temporal Link Prediction via Temporal Walk Matrix Projection (NeurIPS 2024 · similarity 0.86)
- Using Time-Aware Graph Neural Networks to Predict Temporal Centralities in Dynamic Graphs (NeurIPS 2024 · similarity 0.86)
- MAGE: Model-Level Graph Neural Networks Explanations via Motif-based Graph Generation (ICLR 2025 · similarity 0.86)
Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.
