1 paper
Peng Ding, Jun Kuang, Wen Sun +5
Large language models (LLMs) remain vulnerable to jailbreaking attacks despite their impressive capabilities. Investigating these weaknesses is crucial for robust safety mechanisms…