ChatBug: A Common Vulnerability of Aligned LLMs Induced by Chat Templates

Documenti analoghi

• SafeDecoding: Defending against Jailbreak Attacks via Safety-Aware Decoding
  di: Xu, Zhangchen, et al.
  Pubblicazione: (2024)
• ACE: A Model Poisoning Attack on Contribution Evaluation Methods in Federated Learning
  di: Xu, Zhangchen, et al.
  Pubblicazione: (2024)
• CleanGen: Mitigating Backdoor Attacks for Generation Tasks in Large Language Models
  di: Li, Yuetai, et al.
  Pubblicazione: (2024)
• Brave: Byzantine-Resilient and Privacy-Preserving Peer-to-Peer Federated Learning
  di: Xu, Zhangchen, et al.
  Pubblicazione: (2024)
• SoSBench: Benchmarking Safety Alignment on Six Scientific Domains
  di: Jiang, Fengqing, et al.
  Pubblicazione: (2025)