Xuehai Tang

jailbreak attack (4) large language model (4) safety alignment (2) chain-of-thought reasoning (1) model safety (1) model adaptation (1) model editing (1) constrained optimization (1) model alignment (1) backdoor attack (1) adversarial attack (1) adversarial defense (1) lyapunov optimization (1) low-rank adaptation (1) red teaming (1) cognitive bia (1) knowledge preservation (1) vulnerability identification (1) attack success rate (1) sequential editing (1)