GESR: Graph-Based Edge Semantic Reconstruction for Stealthy Communication Detection with Benign-Only Training

Documenti analoghi

• Graph-Aware Stealthy Poison-Text Backdoors for Text-Attributed Graphs
  di: Luo, Qi, et al.
  Pubblicazione: (2026)
• Stealthy Yet Effective: Distribution-Preserving Backdoor Attacks on Graph Classification
  di: Wang, Xiaobao, et al.
  Pubblicazione: (2025)
• The 'Sure' Trap: Multi-Scale Poisoning Analysis of Stealthy Compliance-Only Backdoors in Fine-Tuned Large Language Models
  di: Tan, Yuting, et al.
  Pubblicazione: (2025)
• Lurking in the shadows: Unveiling Stealthy Backdoor Attacks against Personalized Federated Learning
  di: Lyu, Xiaoting, et al.
  Pubblicazione: (2024)
• The Surprising Harmfulness of Benign Overfitting for Adversarial Robustness
  di: Hao, Yifan, et al.
  Pubblicazione: (2024)