跳到正文
arXiv cs.AI· Stavros Damianakis, Konstantinos Psounis·· 2 小时前AI 评分53

AutoPI 提出分类法引导的自适应提示注入红队方法

AutoPI: Taxonomy-Guided Adaptive Red Teaming for Prompt Injection in LLM Agents

AI 导读

研究者提出 AutoPI,一种面向 LLM 智能体提示注入的自适应红队方法,用 PIAtlas 分类法引导攻击搜索并以技术覆盖度衡量效果,而非只追求攻击成功率。

正文 · 原文

View PDF HTML (experimental)

Abstract:Prompt injection is the dominant unsolved security risk for LLM agents, and the community now agrees that robustness must be measured against adaptive attackers rather than fixed benchmark injections. Many adaptive red teamers exist, and most are built to maximize attack success rate (ASR). Yet ASR counts how often a target breaks, not in how many different ways. A red teamer can achieve a high ASR by reusing a few techniques that work, leaving the others untested and absent from the attacks used to harden models. We present AutoPI, an adaptive red teamer designed to find many different ways to break a target, not just to break it often. AutoPI uses PIAtlas, a taxonomy of prompt-injection techniques, both to guide its search and to measure coverage. For each scenario, an attacker LLM writes an injection for every sub-technique in the taxonomy and then reworks the ones that fail using feedback from an LLM judge. The method is black-box and training-free, it plugs into existing benchmarks in place of their fixed attack templates, and it stays up to date as the taxonomy evolves. We evaluate AutoPI on AgentDojo and IPI Arena against open-weight and proprietary target models. Alongside ASR, we report technique coverage, the fraction of taxonomy classes, techniques, and sub-techniques that yield at least one successful injection. On AgentDojo, AutoPI finds a successful injection against Qwen3.6-35B in all six classes of the taxonomy and for more than 90% of its techniques. Against Claude Haiku 5.5, the most robust target we evaluate, it achieves between 20% and 50% higher ASR and between 25% and 400% higher class-level coverage than recent state-of-the-art adaptive red teamers, while requiring no training. We release our code and the generated attacks to support reproducible, coverage-aware evaluation.
Comments: 18 pages, 8 figures, 5 tables. Project page at this https URL and code at this https URL
Subjects: Cryptography and Security (cs.CR); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Cite as: arXiv:2611.24313 [cs.CR]
  (or arXiv:2611.24313v1 [cs.CR] for this version)
  https://doi.org/10.48550/arXiv.2611.24313

arXiv-issued DOI via DataCite (pending registration)

Submission history

From: Stavros Damianakis [view email]
[v1] Wed, 18 Nov 2026 12:20:58 UTC (97 KB)

来源:arXiv cs.AI · arxiv.org