Agent-SafetyBench

档案区分来源方法阅读与既有目录信息；已经公开不代表完成独立证据复核或双语编辑复核。

测量构念

智能体与环境、工具交互时有哪些安全故障？

任务与样本

349环境、2,000案例覆盖八风险类及十故障模式，细化、去重并统一既有案例。

判分与失效情形

案例安全分按总分及故障模式汇总，帮助性另测。

系统条件

349环境、2,000案例、16智能体。

统计含义

来源报告的结果表述；这不等同于已确定运行次数、分母或估计方法： 16个受测智能体总安全分均未超60%。 防御提示收益有限，部分弱模型没有改善（§4.5）。

效度与解释

构造场景及特定判分不等于部署事故率，高能力也不保证安全。

适用研究问题

模型建议：查看故障模式与帮助性明细，不只看总分。

• Agent-SafetyBench : Evaluating the Safety of LLM Agents · AbstractSelected original resource https://arxiv.org/html/2412.14470; sha256:d75e5fb230a89a9ffa0484fbfe694ac609b7b5de3aa59204cadf73b473737338; archived source capture; exact capture time not independently established
• Agent-SafetyBench : Evaluating the Safety of LLM Agents · §3.2.2 Refine Existing DatasetsSelected original resource https://arxiv.org/html/2412.14470; sha256:d75e5fb230a89a9ffa0484fbfe694ac609b7b5de3aa59204cadf73b473737338; archived source capture; exact capture time not independently established
• Agent-SafetyBench : Evaluating the Safety of LLM Agents · §4.2 Main ResultsSelected original resource https://arxiv.org/html/2412.14470; sha256:d75e5fb230a89a9ffa0484fbfe694ac609b7b5de3aa59204cadf73b473737338; archived source capture; exact capture time not independently established
• Agent-SafetyBench : Evaluating the Safety of LLM Agents · §4.5 Discussion on DefenseSelected original resource https://arxiv.org/html/2412.14470; sha256:d75e5fb230a89a9ffa0484fbfe694ac609b7b5de3aa59204cadf73b473737338; archived source capture; exact capture time not independently established