Labels Predicted by AI
インダイレクトプロンプトインジェクション 注意メカニズム 大規模言語モデル
Please note that these labels were automatically added by AI. Therefore, they may not be entirely accurate.
For more details, please see the About the Literature Database page.
Abstract
Large Language Models (LLMs) have revolutionized various domains but remain vulnerable to prompt injection attacks, where malicious inputs manipulate the model into ignoring original instructions and executing designated action. In this paper, we investigate the underlying mechanisms of these attacks by analyzing the attention patterns within LLMs. We introduce the concept of the distraction effect, where specific attention heads, termed important heads, shift focus from the original instruction to the injected instruction. Building on this discovery, we propose Attention Tracker, a training-free detection method that tracks attention patterns on instruction to detect prompt injection attacks without the need for additional LLM inference. Our method generalizes effectively across diverse models, datasets, and attack types, showing an AUROC improvement of up to 10.0 small LLMs. We demonstrate the robustness of our approach through extensive evaluations and provide insights into safeguarding LLM-integrated systems from prompt injection vulnerabilities.