Research map / Graph learning / GNN architectures
Combating Bilateral Edge Noise for Robust Link Prediction
Zhanke Zhou, Jiangchao Yao, Jiaxu Liu, Xiawei Guo, quanming yao, LI He, Liang Wang, Bo Zheng, Bo Han
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 |
|---|---|---|---|
| remove hybrid augmentation (ablation) | RGIB-SSL (SAGE) · Cora, bilateral noise 60% | AUC: .8596 -> .8150 | Improved |
| remove self-adversarial alignment loss (ablation) | RGIB-SSL (SAGE) · Cora, bilateral noise 60% | AUC: .8596 -> .8410 | Improved |
| remove label supervision (lambda_s=0) (ablation) | RGIB-SSL (SAGE) · Cora, bilateral noise 60% | AUC: .8596 -> .7480 | Improved |
| remove alignment term (lambda_a=0) (ablation) | RGIB-SSL (SAGE) · Cora, bilateral noise 60% | AUC: .8596 -> .8194 | Improved |
| remove uniformity term (lambda_u=0) (ablation) | RGIB-SSL (SAGE) · Cora, bilateral noise 60% | AUC: .8596 -> .8355 | Improved |
| remove edge selection (ablation) | RGIB-REP (SAGE) · Cora, bilateral noise 60% | AUC: .7611 -> .7515 | Improved |
| remove label selection (ablation) | RGIB-REP (SAGE) · Cora, label noise 60% | AUC: .8095 -> .7847 | Improved |
| remove topology constraint (lambda_A=0) (ablation) | RGIB-REP (SAGE) · Cora, input noise 60% | AUC: .8487 -> .7699 | Improved |
| remove label constraint (lambda_Y=0) (ablation) | RGIB-REP (SAGE) · Cora, input noise 60% | AUC: .8487 -> .8106 | Improved |
| remove hybrid augmentation (ablation) | RGIB-SSL (SAGE) · Chameleon, input noise 60% | AUC: .9758 -> .9746 | No clear effect |
| replace GCN standard training with RGIB-SSL under bilateral noise | GCN · Cora, bilateral noise 40% | AUC: .7419 -> .8554 | Improved |
| apply existing robust methods (VIB, SupCon) to bilateral noise | GCN · Cora, bilateral noise 40% | AUC: .7419 -> .7810 | Mixed |
Closest papers by meaning
- Towards Understanding and Reducing Graph Structural Noise for GNNs (ICML 2023 · similarity 0.89)
- Training Robust Graph Neural Networks by Modeling Noise Dependencies (NeurIPS 2025 · similarity 0.89)
- Implicit degree bias in the link prediction task (ICML 2025 · similarity 0.86)
- NoisyGL: A Comprehensive Benchmark for Graph Neural Networks under Label Noise (NeurIPS 2024 · similarity 0.86)
- A Topological Perspective on Demystifying GNN-Based Link Prediction Performance (ICLR 2024 · similarity 0.86)
- Robust Node Classification on Graph Data with Graph and Label Noise (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.
