Operationalizing a Threat Model for Red-Teaming Large Language Models (LLMs)
This work addresses security vulnerabilities in LLM-based systems for practitioners, but it is incremental as it synthesizes existing research rather than introducing new methods.
The paper tackles the problem of securing large language models (LLMs) by developing a threat model and systematizing knowledge on red-teaming attacks, resulting in a taxonomy and framework to improve security and robustness.
Creating secure and resilient applications with large language models (LLM) requires anticipating, adjusting to, and countering unforeseen threats. Red-teaming has emerged as a critical technique for identifying vulnerabilities in real-world LLM implementations. This paper presents a detailed threat model and provides a systematization of knowledge (SoK) of red-teaming attacks on LLMs. We develop a taxonomy of attacks based on the stages of the LLM development and deployment process and extract various insights from previous research. In addition, we compile methods for defense and practical red-teaming strategies for practitioners. By delineating prominent attack motifs and shedding light on various entry points, this paper provides a framework for improving the security and robustness of LLM-based systems.