Research map / Graph learning / Graph self-supervised learning
RABot: Reinforcement-Guided Graph Augmentation for Imbalanced and Noisy Social Bot Detection
Longlong Zhang, Xi Wang, Haotong Du, Yangyi Xu, Zhuo Liu, Yang Liu
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 multi-head attention fusion in node encoder (ablation) | RABot (RGT) · Twibot-20 | accuracy: 87.40 -> 87.92 | Improved |
| add neighborhood-aware feature augmentation (ablation) | RABot (RGT) · MGTAB | accuracy: 89.90 -> 90.74 | Improved |
| add neighborhood-aware feature augmentation (ablation) | RABot (RGCN) · MGTAB | accuracy: 89.88 -> 91.16 | Improved |
| add RL edge filtering (ablation) | RABot (RGT) · Twibot-20 | accuracy: 86.96 -> 87.92 | Improved |
| add RL edge filtering (ablation) | RABot (RGCN) · MGTAB | accuracy: 89.73 -> 91.16 | Improved |
| add GNN classification module (ablation) | RABot (RGT) · MGTAB | accuracy: 87.23 -> 90.74 | Improved |
| use RL-adaptive threshold instead of best fixed threshold (ablation) | RABot (RGCN) · Twibot-20 | accuracy: 86.73 -> 87.36 | Improved |
| use RL-adaptive threshold instead of best fixed threshold (ablation) | RABot (RGCN) · MGTAB | accuracy: 90.54 -> 91.16 | Improved |
| plug RABot into GCN backbone | GCN · Twibot-20 | accuracy: 77.62 -> 82.08 | Improved |
| plug RABot into RGT backbone | RGT · Cresci-15 | accuracy: 97.13 -> 99.14 | Improved |
| plug RABot into RGCN backbone | BotRGCN · MGTAB | accuracy: 89.27 -> 91.16 | Improved |
| vary loss weights lambda_s and lambda_e (ablation) | RABot (GAT) · Twibot-20 and MGTAB | accuracy: <1% accuracy variation | No clear effect |
Closest papers by meaning
- Enhancing Robustness of Graph Neural Networks on Social Media with Explainable Inverse Reinforcement Learning (NeurIPS 2024 · similarity 0.87)
- Adversarial Socialbots Modeling Based on Structural Information Principles (AAAI 2024 · similarity 0.86)
- Grimm: A Plug-and-Play Perturbation Rectifier for Graph Neural Networks Defending Against Poisoning Attacks (AAAI 2025 · similarity 0.83)
- Partitioning Message Passing for Graph Fraud Detection (ICLR 2024 · similarity 0.83)
- $\texttt{G1}$: Teaching LLMs to Reason on Graphs with Reinforcement Learning (NeurIPS 2025 · similarity 0.83)
- Towards Effective and General Graph Unlearning via Mutual Evolution (AAAI 2024 · similarity 0.83)
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.
