Research map / Label-efficient and robust learning / Knowledge distillation
Generalist Graph Anomaly Detection via Prototype-Based Distillation
Yiming Xu, Zihan Chen, Zhen Peng, Song Wang, Bin Shi, Bo Dong, Chao Shen
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 prototype-driven soft-label distillation (ablation) | ProMoS · 6 graphs avg | AUROC: 56.32 -> 75.46 | Improved |
| add discrepancy-aware commitment and refinement loss (ablation) | ProMoS · 6 graphs avg | AUROC: 70.20 -> 75.46 | Improved |
| use a pre-trained SSL teacher instead of random GCN (ablation) | ProMoS · 6 graphs avg | AUROC: 69.93 -> 75.46 | Improved |
| alternate prototype and teacher updates instead of joint optimization (ablation) | ProMoS · 6 graphs avg | AUROC: 75.46 vs 75.32 / 75.03 | No clear effect |
| swap GCA teacher for GraphCL (ablation) | ProMoS · 6 graphs avg | AUROC: 80.92 -> 79.45 | Mixed |
| swap GCA teacher for GraphMAE (ablation) | ProMoS · 6 graphs avg | AUROC: 80.92 -> 77.81 | Hurt |
| swap GCA teacher for DGI (ablation) | ProMoS · 6 graphs avg | AUROC: 80.92 -> 80.01 | Hurt |
| use unsupervised distillation from SSL teacher vs supervised generalist methods | ProMoS · Cora | AUROC: 66.53 -> 84.56 | Improved |
| ProMoS vs DOMINANT on Weibo | ProMoS · Weibo | AUROC: 92.88 -> 91.74 | Hurt |
| ProMoS efficiency vs GCN | ProMoS · training/inference time | speedup: 4.8x faster training, 1.4x faster inference | Improved |
Closest papers by meaning
- ARC: A Generalist Graph Anomaly Detector with In-Context Learning (NeurIPS 2024 · similarity 0.92)
- Graph Anomaly Detection as Dynamical Transport: Training-Free Scoring via Empirical Bayes (NeurIPS 2026 · similarity 0.91)
- UniFORM: Towards Unified Framework for Anomaly Detection on Graphs (AAAI 2025 · similarity 0.91)
- Leveraging Diffusion Model as Pseudo-Anomalous Graph Generator for Graph-Level Anomaly Detection (ICML 2025 · similarity 0.91)
- Open-Set Graph Anomaly Detection via Normal Structure Regularisation (ICLR 2025 · similarity 0.90)
- Generative Semi-supervised Graph Anomaly Detection (NeurIPS 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.
