The Instruction Hierarchy: Training LLMs to Prioritize Privileged Instructions
arxiv(2024)
摘要
Today's LLMs are susceptible to prompt injections, jailbreaks, and other
attacks that allow adversaries to overwrite a model's original instructions
with their own malicious prompts. In this work, we argue that one of the
primary vulnerabilities underlying these attacks is that LLMs often consider
system prompts (e.g., text from an application developer) to be the same
priority as text from untrusted users and third parties. To address this, we
propose an instruction hierarchy that explicitly defines how models should
behave when instructions of different priorities conflict. We then propose a
data generation method to demonstrate this hierarchical instruction following
behavior, which teaches LLMs to selectively ignore lower-privileged
instructions. We apply this method to GPT-3.5, showing that it drastically
increases robustness – even for attack types not seen during training – while
imposing minimal degradations on standard capabilities.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要