XGUARD：一个用于评估大语言模型在极端主义内容上安全失效的分级基准
XGUARD: A Graded Benchmark for Evaluating Safety Failures of Large Language Models on Extremist Content
AI总结：
本文提出XGUARD基准，用五级危险等级和攻击严重性曲线评估大语言模型生成极端主义内容的失效严重程度，并通过六种模型和两种防御策略揭示当前安全缺口与鲁棒性-表达自由权衡。
摘要：
大语言模型（LLMs）能够生成从意识形态言论到明确暴力指示的内容。然而，现有的安全评估通常依赖简单的二元标签（安全和不安全），忽视了这些输出所构成的细微风险谱系。为解决这一问题，我们提出了XGUARD，一个旨在评估大语言模型生成的极端主义内容严重程度的基准与评估框架。XGUARD包含3,840个红队提示，这些提示来源于社交媒体和新闻等真实世界数据，涵盖广泛的意识形态化场景。我们的框架将模型响应划分为五个危险等级（0到4），从而能够对失效的频率和严重程度进行更细致的分析。我们引入了可解释的攻击严重性曲线（ASC），用于可视化脆弱性并比较不同威胁强度下的防御机制。利用XGUARD，我们评估了六种流行的大语言模型和两种轻量级防御策略，揭示了当前安全缺口以及鲁棒性与表达自由之间的权衡方面的关键见解。我们的工作强调了分级安全指标对于构建可信大语言模型的价值。
英文摘要：
Large Language Models (LLMs) can generate content spanning ideological rhetoric to explicit instructions for violence. However, existing safety evaluations often rely on simplistic binary labels (safe and unsafe), overlooking the nuanced spectrum of risk these outputs pose. To address this, we present XGUARD, a benchmark and evaluation framework designed to assess the severity of extremist content generated by LLMs. XGUARD includes 3,840 red teaming prompts sourced from real world data such as social media and news, covering a broad range of ideologically charged scenarios. Our framework categorizes model responses into five danger levels (0 to 4), enabling a more nuanced analysis of both the frequency and severity of failures. We introduce the interpretable Attack Severity Curve (ASC) to visualize vulnerabilities and compare defense mechanisms across threat intensities. Using XGUARD, we evaluate six popular LLMs and two lightweight defense strategies, revealing key insights into current safety gaps and trade-offs between robustness and expressive freedom. Our work underscores the value of graded safety metrics for building trustworthy LLMs.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
