PromptBench: Towards Evaluating the Robustness of Large Language Models on Adversarial Prompts

Open in new window