When Safety Becomes a Vulnerability: Exploiting LLM Alignment Homogeneity for Transferable Blocking in RAG

Documenti analoghi

• KEPo: Knowledge Evolution Poison on Graph-based Retrieval-Augmented Generation
  di: Chen, Qizhi, et al.
  Pubblicazione: (2026)
• The Trust Paradox in LLM-Based Multi-Agent Systems: When Collaboration Becomes a Security Vulnerability
  di: Xu, Zijie, et al.
  Pubblicazione: (2025)
• When Safe Models Merge into Danger: Exploiting Latent Vulnerabilities in LLM Fusion
  di: Li, Jiaqing, et al.
  Pubblicazione: (2026)
• When RSA Fails: Exploiting Prime Selection Vulnerabilities in Public Key Cryptography
  di: Nikzad, Murtaza, et al.
  Pubblicazione: (2025)
• Uncovering Logit Suppression Vulnerabilities in LLM Safety Alignment
  di: Li, Yuxi, et al.
  Pubblicazione: (2024)