Improved Few-Shot Jailbreaking Can Circumvent Aligned Language Models and Their Defenses (NeurIPS 2024)
Primary LanguagePythonMIT LicenseMIT
No issues in this repository yet.