Vocabulary Attack to Hijack Large Language Model Applications

Understand

The fast advancements in Large Language Models (LLMs) are driving an increasing number of applications.

• Together with the growing number of users, we also see an increasing number of attackers who try to outsmart these systems.
• They want the model to reveal confidential information, specific false information, or offensive behavior.
• To this end, they manipulate their instructions for the LLM by inserting separators or rephrasing them systematically until they reach their goal.