Disentangling Adversarial Prompts: A Semantic-Graph Defense for Robust LLM Security

Documenti analoghi

• Beyond Vulnerabilities: A Survey of Adversarial Attacks as Both Threats and Defenses in Computer Vision Systems
  di: Guo, Zhongliang, et al.
  Pubblicazione: (2025)
• Adversarial Attacks and Defenses on Text-to-Image Diffusion Models: A Survey
  di: Zhang, Chenyu, et al.
  Pubblicazione: (2024)
• Autoencoder-based Denoising Defense against Adversarial Attacks on Object Detection
  di: Song, Min Geun, et al.
  Pubblicazione: (2025)
• T2I-RiskyPrompt: A Benchmark for Safety Evaluation, Attack, and Defense on Text-to-Image Model
  di: Zhang, Chenyu, et al.
  Pubblicazione: (2025)
• Proactive Adversarial Defense: Harnessing Prompt Tuning in Vision-Language Models to Detect Unseen Backdoored Images
  di: Stein, Kyle, et al.
  Pubblicazione: (2024)