jailbreaking attack

large language model (13587) adversarial attack (1621) model safety (101) adversarial prompt (65) safety alignment (178) prompt optimization (190) red teaming (61) prompt injection (101) multimodal large language model (966) jailbreak attack (225)