XGUARD: A Graded Benchmark for Evaluating Safety Failures of Large Language Models on Extremist Content
This work addresses the need for nuanced safety assessments in LLMs to prevent extremist content generation, though it is incremental as it builds on existing evaluation frameworks.
The authors tackled the problem of simplistic safety evaluations for large language models (LLMs) by introducing XGUARD, a benchmark with 3,840 prompts and a five-level danger categorization, which revealed key safety gaps and trade-offs in six popular LLMs and two defense strategies.
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.