Research map / Adversarial robustness and security / Attacks on text and graphs
Towards Robust Text-Attributed Federated Graph Learning: Multimodal Threats and Defense
Zitong Shi, Guancheng Wan, Wenke Huang, Yuxin Wu, Quan Zhang, Mang Ye
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 text attack to structure attack (GTAE) (ablation) | GTAE · PubMed, LLaGA | accuracy under attack (%): 68.19 -> 48.59 | Improved |
| add structure attack to text attack (GTAE) (ablation) | GTAE · PubMed, LLaGA | accuracy under attack (%): 53.79 -> 48.59 | Improved |
| add structure attack to text attack (GTAE) (ablation) | GTAE · Cora, LLaGA | accuracy under attack (%): 67.31 -> 57.34 | Improved |
| add text defense to structure defense (STRUM) (ablation) | STRUM · PubMed under GTAE | accuracy (%): 56.42 -> 84.49 | Improved |
| add structure defense to text defense (STRUM) (ablation) | STRUM · PubMed under GTAE | accuracy (%): 77.15 -> 84.49 | Improved |
| add both defense modules (STRUM) (ablation) | LLaGA · Cora under GTAE | accuracy (%): 57.34 -> 81.47 | Improved |
| use GTAE instead of the strongest baseline attack | LLaGA · PubMed | attack success rate (%): 37.62 | Improved |
| use STRUM instead of SAFER | LLaGA · PubMed under GTAE | accuracy (%): 64.12 -> 84.49 | Improved |
| use STRUM on GraphPrompter backbone instead of SAFER | GraphPrompter · Cora under GTAE | accuracy (%): 56.83 -> 61.47 | Improved |
| apply structure defenses ProGNN or SimPGCN alone | GraphPrompter · Cora under GTAE | accuracy (%): 51.90 -> 51.37 | Hurt |
| increase structure exposure rate of attack from 0 to 0.15 | GTAE · PubMed, GraphTranslator | accuracy (%): 64.12 -> 48.35 | Hurt |
| set structure exposure rate 0 and text perturbation 0.10 in defense | STRUM · PubMed, GraphTranslator | accuracy (%): 61.9 | Improved |
Earlier methods this paper compares with
Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.
- TextHoaxer — TextHoaxer: Budgeted Hard-Label Adversarial Attacks on Text (AAAI 2022) · better on 5 of 5 shared results
Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.
Closest papers by meaning
- Intruding with Words: Towards Understanding Graph Injection Attacks at the Text Level (NeurIPS 2024 · similarity 0.89)
- Generalization Principles for Inference over Text-Attributed Graphs with Large Language Models (ICML 2025 · similarity 0.86)
- SSTAG: Structure-Aware Self-Supervised Learning Method for Text-Attributed Graphs (NeurIPS 2025 · similarity 0.86)
- Beyond the Golden Teacher: Enhancing Graph Learning through LLM-GNN Co-teaching (NeurIPS 2026 · similarity 0.84)
- Dynamic Bundling with Large Language Models for Zero-Shot Inference on Text-Attributed Graphs (NeurIPS 2025 · similarity 0.84)
- DTGB: A Comprehensive Benchmark for Dynamic Text-Attributed Graphs (NeurIPS 2024 · similarity 0.84)
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.
