Research map / Graph learning / GNN architectures
Towards Multi-Grained Explainability for Graph Neural Networks
Xiang Wang, Yingxin Wu, An Zhang, Xiangnan He, Tat-seng Chua
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 |
|---|---|---|---|
| add contrastive loss to pre-training (ablation) | ReFine-FT · Mutagenicity | ACC-AUC: 0.888 -> 0.945 | Improved |
| add contrastive loss to pre-training (ablation) | ReFine-FT · VG-5 | ACC-AUC: 0.891 -> 0.906 | Improved |
| add contrastive loss to pre-training (ablation) | ReFine-FT · MNIST | ACC-AUC: 0.526 -> 0.587 | Improved |
| add contrastive loss to pre-training (ablation) | ReFine-FT · BA-3motif | ACC-AUC: 0.610 -> 0.616 | Improved |
| add fine-tuning phase with selection module (ablation) | ReFine · MNIST | ACC@0.4: 41.4% -> 71.4% | Improved |
| add fine-tuning phase with selection module (ablation) | ReFine · Mutagenicity | ACC-AUC: 0.945 -> 0.955 | Improved |
| add fine-tuning phase with selection module (ablation) | ReFine · VG-5 | ACC-AUC: 0.906 -> 0.914 | Improved |
| add fine-tuning phase with selection module (ablation) | ReFine · BA-3motif | ACC@0.6: 65.7% -> 72.8% | Improved |
| add fine-tuning phase with selection module (ablation) | ReFine · BA-3motif | Recall@5: 0.299 -> 0.304 | Improved |
| class-wise attributors plus contrastive learning vs PGExplainer single attributor | ReFine-FT · MNIST | ACC-AUC: 0.587 vs PGM-Explainer 0.615 | Mixed |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- PGM-Explainer — PGM-Explainer: Probabilistic Graphical Model Explanations for Graph Neural Networks (NeurIPS 2020) · better on 5 of 5 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
- Towards Robust Fidelity for Evaluating Explainability of Graph Neural Networks (ICLR 2024 · similarity 0.91)
- Beyond Topological Self-Explainable GNNs: A Formal Explainability Perspective (ICML 2025 · similarity 0.91)
- Self-Interpretable Graph Learning with Sufficient and Necessary Explanations (AAAI 2024 · similarity 0.91)
- Factorized Explainer for Graph Neural Networks (AAAI 2024 · similarity 0.90)
- Stratified GNN Explanations through Sufficient Expansion (AAAI 2024 · similarity 0.90)
- Generating In-Distribution Proxy Graphs for Explaining Graph Neural Networks (ICML 2024 · similarity 0.90)
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.
