PANDAS: Improving Many-shot Jailbreaking via Positive Affirmation, Negative Demonstration, and Adaptive Sampling

Documenti analoghi

• Improving Adversarial Transferability via Model Alignment
  di: Ma, Avery, et al.
  Pubblicazione: (2023)
• Improved Techniques for Optimization-Based Jailbreaking on Large Language Models
  di: Jia, Xiaojun, et al.
  Pubblicazione: (2024)
• Jailbreak and Guard Aligned Language Models with Only Few In-Context Demonstrations
  di: Wei, Zeming, et al.
  Pubblicazione: (2023)
• Deciphering the Chaos: Enhancing Jailbreak Attacks via Adversarial Prompt Translation
  di: Li, Qizhang, et al.
  Pubblicazione: (2024)
• A StrongREJECT for Empty Jailbreaks
  di: Souly, Alexandra, et al.
  Pubblicazione: (2024)