Research map / Adversarial robustness and security / Attacks on text and graphs
Robust Heterogeneous Graph Neural Networks against Adversarial Attacks
Read the paperCheck your idea against 108,967 papers
AI-extracted summary. Written by AI from the paper's title only (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 the RoHe attention purifier to HAN | HAN · ACM | Micro-F1: 0.528 -> 0.904 | Helped |
| Add the RoHe attention purifier to HAN | HAN · DBLP | Micro-F1: 0.332 -> 0.936 | Helped |
| Add the RoHe attention purifier to HAN | HAN · Aminer | Micro-F1: 0.346 -> 0.840 | Helped |
| Add the RoHe attention purifier to MAGNN | MAGNN · ACM | Micro-F1: 0.711 -> 0.901 | Helped |
| Add the RoHe attention purifier to MAGNN | MAGNN · DBLP | Micro-F1: 0.620 -> 0.798 | Helped |
| Add pruning without the transiting-probability prior (ablation) | HAN · ACM | Micro-F1: 0.780 -> 0.904 | Hurt |
| Add pruning without the transiting-probability prior (ablation) | HAN · DBLP | Micro-F1: 0.686 -> 0.936 | Hurt |
| Add the transiting-probability prior without pruning (ablation) | HAN · ACM | Micro-F1: 0.900 -> 0.904 | No clear effect |
| Add the transiting-probability prior without pruning (ablation) | HAN · DBLP | Micro-F1: 0.760 -> 0.936 | Hurt |
| Add the transiting-probability component only (ablation) | GTN · ACM | Micro-F1: 0.786 -> 0.892 | Helped |
| Add the transiting-probability component only (ablation) | GTN · DBLP | Micro-F1: 0.564 -> 0.644 | Helped |
Closest papers by meaning
- Bounding the Expected Robustness of Graph Neural Networks Subject to Node Feature Attacks (ICLR 2024 · similarity 0.89)
- A Simple and Yet Fairly Effective Defense for Graph Neural Networks (AAAI 2024 · similarity 0.89)
- Robust Graph Neural Networks via Unbiased Aggregation (NeurIPS 2024 · similarity 0.88)
- Refining Latent Homophilic Structures over Heterophilic Graphs for Robust Graph Convolution Networks (AAAI 2024 · similarity 0.88)
- Graph Agent Network: Empowering Nodes with Inference Capabilities for Adversarial Resilience (AAAI 2025 · similarity 0.87)
- Value at Adversarial Risk: A Graph Defense Strategy against Cost-Aware Attacks (AAAI 2024 · similarity 0.87)
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.