1
Capability Minimization as a Safety Primitive: Risk-Aware Causal Gating for Least-Privilege LLM Agents
这篇论文提出了一种名为“能力最小化”的安全原语，通过风险感知的因果门控机制实现最小权限LLM Agent，为AI安全提供新思路。
arXiv:2606.13884v1 Announce Type: new Abstract: Modern decision systems increasingly rely on learned components whose outputs may be confident yet wro…