AI Paper Atlas

Research map / Graph learning / Graph self-supervised learning

Towards Self-Interpretable Graph-Level Anomaly Detection

Yixin Liu, Kaize Ding, Qinghua Lu, Fuyi Li, Leo Yu Zhang, Shirui Pan

NeurIPS 2023 · Graph learning › Graph self-supervised learning · approach: Graph anomaly detection · 24 citations (OpenAlex)

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 changedWhereMeasuredEffect
Use dual-hypergraph views rather than augmentation-based views (ablation)SIGNET · BM-MSAD-AUC: 63.26 -> 94.01Improved
Use dual-hypergraph views rather than augmentation-based views (ablation)SIGNET · MNIST-0AD-AUC: 48.75 -> 83.25Improved
Use dual-hypergraph views rather than structural-property-based views (ablation)SIGNET · BM-MSAD-AUC: 92.79 -> 94.01Improved
Use dual-hypergraph views rather than structural-property-based views (ablation)SIGNET · MNIST-0AD-AUC: 55.29 -> 83.25Improved
Use one shared extractor instead of two extractors with SKL divergence (ablation)SIGNET · BM-MSAD-AUC: 92.44 -> 94.01Improved
Use one shared extractor instead of two extractors with SKL divergence (ablation)SIGNET · MNIST-0AD-AUC: 81.77 -> 83.25Improved
Use InfoNCE rather than the JS MI estimator (ablation)SIGNET · BM-MSAD-AUC: 75.21 -> 94.01Improved
Use InfoNCE rather than the JS MI estimator (ablation)SIGNET · MNIST-0AD-AUC: 71.78 -> 83.25Improved
Use InfoNCE rather than the DV MI estimator (ablation)SIGNET · BM-MSAD-AUC: 74.87 -> 94.01Improved
Use InfoNCE rather than the DV MI estimator (ablation)SIGNET · MNIST-0AD-AUC: 70.31 -> 83.25Improved
Apply SIGNET instead of OCGTLBM-MN · BM-MNAD-AUC: 99.49 -> 93.41Hurt
Apply SIGNET instead of OCGINMNIST-1 · MNIST-1AD-AUC: 98.25 -> 90.12Hurt

Closest papers by meaning

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.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy