Large Language Models

Jailbreaking LLMs Without Gradients or Priors: Effective Transferable Attacks via Random Local Search
Zhakshylyk Nurlanov, Frank R. Schmidt, Florian Bernard. Preprint 2025
Zhakshylyk Nurlanov, Frank R. Schmidt, Florian Bernard
Cite
Jailbreaking LLMs Without Gradients or Priors: Effective Transferable Attacks via Random Local Search