arXiv cs.AI· Tianyu Chen, Chujia Hu, Wenjie Wang·· 9 天前AI 评分29
Safety Sentry：通过 EXECUTE-ASK-REFUSE 路由实现上下文感知的人类干预
SAFETY SENTRY: Context-Aware Human Intervention via EXECUTE-ASK-REFUSE Routing
AI 导读
Safety Sentry 将 LLM 智能体的安全防护重构为针对每个实例的 {EXECUTE, ASK, REFUSE} 三路路由决策，用轻量级 guard 模型实现，推理仅需一次解码调用。单个解码时阈值即可让同一 checkpoint 在不同风险容忍度的部署中重新定位，无需重新训练。该模型在整体准确率和安全相关召回率上优于多种开放权重与前沿闭源基线，同时控制双向错误率。
正文 · 原文
Abstract:LLM agents act on real-world environments through tool calls, and a single misjudged action can cause irreversible harm. The standard safeguard is a guard model that labels each proposed action as safe or unsafe, but this binary view conflates two distinct decisions: whether the action is harmful in itself, and whether it is appropriate given the user's context. It also operates at the granularity of action categories rather than individual instances, producing routine interruptions that erode autonomy and train users to wave through the most consequential alerts. We reframe the problem as a per-instance three-way routing decision over {EXECUTE, ASK, REFUSE} and instantiate it with Safety Sentry, a lightweight guard model whose inference reduces to a single decoding call. A single decoding-time threshold lets one fixed checkpoint be re-positioned across deployments of differing risk tolerance without retraining. Safety Sentry outperforms a broad set of open-weight and frontier closed-source baselines on overall accuracy and safety-related recall, while controlling both directional error rates this http URL and data are available at this https URL
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2607.13594 [cs.AI] |
| (or arXiv:2607.13594v2 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2607.13594 arXiv-issued DOI via DataCite |
Submission history
From: Chujia Hu [view email]
[v1]
Wed, 15 Jul 2026 08:38:16 UTC (1,220 KB)
[v2]
Thu, 8 Oct 2026 06:32:32 UTC (1,042 KB)
来源：arXiv cs.AI · arxiv.org