Jailbreaking Leading Safety-Aligned LLMs with Simple Adaptive Attacks

2025-01-22 · ICLR 2025 Poster · anchor

Findings