Jaturong Kongmanee

Guarding LLMs With a Layered Prompt Injection Representation

TrendAI™ Research has developed a model training procedure for learning an essential representation of prompt injection attacks. The resulting prompt representation exhibits approximately linear separability, allowing the specialized, small-scale classifier trained on features derived from the representation to achieve high classification performance.

Jaturong Kongmanee, Smile Thanapattheerakul