Zhangchen Xu

large language model (8) safety alignment (3) jailbreak attack (3) adversarial attack (2) instruction tuning (2) adversarial learning (2) decoding strategy (2) llm safety (2) knowledge distillation (2) chain-of-thought reasoning (2) code generation (1) backdoor attack (1) text generation (1) synthetic datum (1) harmful content (1) adversarial prompt (1) model safety (1) safety evaluation (1) supervised fine-tuning (1) adversarial training (1)