Research map / Graph learning / GNN architectures
Understanding and Improving Graph Injection Attack by Promoting Unnoticeability
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 HAO to GIA (ablation) | GIA attacks against homophily defenders · Cora, Citeseer, Computers, and Arxiv, non-targeted | maximum test accuracy: 83.07 -> 42.18; 74.72 -> 61.08; 86.03 -> 80.86; 66.95 -> 45.59 | Helped |
| Add HAO to GIA (ablation) | GIA attacks against homophily defenders · Computers, Arxiv, Aminer, and Reddit, targeted | maximum test accuracy: 87.85 -> 72.00; 66.12 -> 38.28; 50.87 -> 22.50; 91.25 -> 64.09 | Helped |
| Add HAO to GIA (ablation) | GIA attacks against defenses overall · Non-targeted and targeted benchmarks | attack performance: up to 30%; up to 15% | Mixed |
| Add HAO to TDGIA (ablation) | TDGIA · Non-targeted and targeted benchmarks | maximum test accuracy: 85.72 -> 79.45; 75.26 -> 72.51; 91.40 -> 90.42; 70.97 -> 63.57; 91.56 -> 77.15; 71.06 -> 42.53; 54.82 -> 29.94; 93.62 -> | Helped |
| Use adaptive heuristic injection instead of TDGIA with HAO | GIA + HAO · Non-targeted and targeted benchmarks | maximum test accuracy | Mixed |
| Increase the HAO regularization weight (ablation) | GIA + HAO · GCN versus homophily defenders | maximum test accuracy: up to 5% | Mixed |
| Use HAO to preserve the homophily distribution | GIA · Graph injection attacks | homophily distribution shift | Helped |
| Use GIA rather than GMA under the stated theorem conditions | Graph adversarial attack · No defense, linearized GNN | attack loss | Helped |
| Apply homophily defenders to GIA without HAO | GIA · Homophily-defender evaluation | attack effectiveness | Hurt |
Closest papers by meaning
- Stronger and Transferable Node Injection Attacks (AAAI 2024 · similarity 0.89)
- Highly Imperceptible Black-Box Graph Injection Attacks with Reinforcement Learning (AAAI 2025 · similarity 0.88)
- Intruding with Words: Towards Understanding Graph Injection Attacks at the Text Level (NeurIPS 2024 · similarity 0.87)
- HyperNear: Unnoticeable Node Injection Attacks on Hypergraph Neural Networks (ICML 2025 · similarity 0.87)
- Are Your Models Still Fair? Fairness Attacks on Graph Neural Networks via Node Injections (NeurIPS 2024 · similarity 0.86)
- Value at Adversarial Risk: A Graph Defense Strategy against Cost-Aware Attacks (AAAI 2024 · similarity 0.86)
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.