Yichuan Mo

adversarial robustness (2) adversarial training (2) image generation (1) model robustness (1) model architecture (1) model safety (1) safety alignment (1) adversarial defense (1) latent space (1) generative adversarial network (1) jailbreaking attack (1) prompt optimization (1) pretrained model (1) mode collapse (1) llm security (1) large language model (1) adversarial tuning (1) gradient masking (1) adversarial learning (1) prompt adversarial tuning (1)