In-Context Learning Can Re-learn Forbidden Tasks

Documenti analoghi

• A Generative Approach to LLM Harmfulness Mitigation with Red Flag Tokens
  di: Dobre, David, et al.
  Pubblicazione: (2025)
• Efficient Adversarial Training in LLMs with Continuous Attacks
  di: Xhonneux, Sophie, et al.
  Pubblicazione: (2024)
• LLM-Safety Evaluations Lack Robustness
  di: Beyer, Tim, et al.
  Pubblicazione: (2025)
• Learning diverse attacks on large language models for robust red-teaming and safety tuning
  di: Lee, Seanie, et al.
  Pubblicazione: (2024)
• Soft Prompt Threats: Attacking Safety Alignment and Unlearning in Open-Source LLMs through the Embedding Space
  di: Schwinn, Leo, et al.
  Pubblicazione: (2024)