1 paper
Yu Pan, Wenlong Yu, Tiejun Wu +4
Large language models (LLMs) have demonstrated remarkable capabilities in complex reasoning tasks. However, they remain highly susceptible to jailbreak attacks that undermine their…