Research map / Graph learning / Graph self-supervised learning
Breaking One-Size-Fits-All: Revisiting Out-of-Distribution Detection on Graphs Under Diverse Distribution Shifts
Chuancheng Song, Hanyang Shen, Yan Dong, Xixun Lin, Yanmin Shang, Yanan Cao
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 |
|---|---|---|---|
| use UniGOD instead of SpaceGNN | graph OOD detection · Average over all datasets | AUROC: 0.8713 -> 0.9016 | Improved |
| add GeoUP adaptive curvature learning (ablation) | UniGOD · Ogbn-ArXiv | AUROC | Improved |
| add DynEVO trajectory uncertainty (ablation) | UniGOD · Ogbn-ArXiv | AUROC | Improved |
| add geometry-dynamics coupling loss (ablation) | UniGOD · Ogbn-ArXiv | AUROC | Improved |
| fuse scores from Euclidean, hyperbolic and hyperspherical spaces (UniGOD-Fuse) (ablation) | UniGOD · Ogbn-ArXiv | AUROC | No clear effect |
| increase curvature excessively beyond moderate positive values (ablation) | kappa-GNN · Ogbn-ArXiv subgraphs | Generalized Fisher Ratio | Hurt |
| use feature-interpolation OOD instead of label leave-out (harder shift needs stronger curvature) | UniGOD · Ogbn-ArXiv subgraphs | Generalized Fisher Ratio: 1.3074 -> 1.2382 | Hurt |
| use kappa-GNN instead of Euclidean GNN at greater depth (ablation) | GNN depth · Ogbn-ArXiv | AUROC | Improved |
| use single hyperbolic space (HGCN) vs adaptive geometry | HGCN · Cora Feature Interpolation | AUROC: 0.6825 vs 0.9256 | Hurt |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.
- SpaceGNN — SpaceGNN: Multi-Space Graph Neural Network for Node Anomaly Detection with Extremely Limited Labels (ICLR 2025) · better on 12 of 12 shared results
- GraphDE — GraphDE: A Generative Framework for Debiased Learning and Out-of-Distribution Detection on Graphs (NeurIPS 2022) · better on 1 of 1 shared result
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
- Graph Out-of-Distribution Detection Goes Neighborhood Shaping (ICML 2024 · similarity 0.91)
- Bridging OOD Detection and Generalization: A Graph-Theoretic View (NeurIPS 2024 · similarity 0.89)
- GOLD: Graph Out-of-Distribution Detection via Implicit Adversarial Latent Generation (ICLR 2025 · similarity 0.89)
- Redundancy-Aware Test-Time Graph Out-of-Distribution Detection (NeurIPS 2025 · similarity 0.89)
- Structural Entropy Guided Unsupervised Graph Out-Of-Distribution Detection (AAAI 2025 · similarity 0.88)
- Harnessing Feature Resonance under Arbitrary Target Alignment for Out-of-Distribution Node Detection (NeurIPS 2025 · similarity 0.88)
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.
