Existing safety evaluation methods for large language models (LLMs) suffer from inherent limitations, including evaluator bias and detection failures arising from model homogeneity, which collectively undermine the robustness of risk evaluation processes. This paper seeks to re-examine the risk evaluation paradigm by introducing a theoretical framework that reconstructs the underlying risk concept space. Specifically, we decompose the latent risk concept space into three mutually exclusive subspaces: the explicit risk subspace (encompassing direct violations of safety guidelines), the implicit risk subspace (capturing potential malicious content that requires contextual reasoning for identification), and the non-risk subspace. Furthermore, we propose RADAR, a multi-agent collaborative evaluation framework that leverages multi-round debate mechanisms through four specialized complementary roles and employs dynamic update mechanisms to achieve self-evolution of risk concept distributions. This approach enables comprehensive coverage of both explicit and implicit risks while mitigating evaluator bias. To validate the effectiveness of our framework, we construct an evaluation dataset comprising 800 challenging cases. Extensive experiments on our challenging testset and public benchmarks demonstrate that RADAR significantly outperforms baseline evaluation methods across multiple dimensions, including accuracy, stability, and self-evaluation risk sensitivity. Notably, RADAR achieves a 28.87% improvement in risk identification accuracy compared to the strongest baseline evaluation method.

翻译：现有大型语言模型（LLM）的安全评估方法存在固有局限性，包括评估者偏见以及由模型同质性导致的检测失效，这些问题共同削弱了风险评估过程的鲁棒性。本文通过引入重构底层风险概念空间的理论框架，旨在重新审视风险评估范式。具体而言，我们将潜在风险概念空间分解为三个互斥的子空间：显性风险子空间（涵盖直接违反安全准则的内容）、隐性风险子空间（捕捉需要上下文推理才能识别的潜在恶意内容）以及非风险子空间。进一步地，我们提出RADAR——一种多智能体协同评估框架，该框架通过四个专业化互补角色利用多轮辩论机制，并采用动态更新机制实现风险概念分布的自我演化。该方法能够全面覆盖显性与隐性风险，同时有效缓解评估者偏见。为验证框架的有效性，我们构建了包含800个挑战性案例的评估数据集。在自建挑战性测试集及公开基准上的大量实验表明，RADAR在准确率、稳定性及自评估风险敏感性等多个维度上显著优于基线评估方法。值得注意的是，相较于最强基线评估方法，RADAR在风险识别准确率上实现了28.87%的提升。