面向智能体5G控制的类型安全决策框架：一种理论驱动的测试平台表征及其适用场景
Type-Safe Decision Frameworks for Agentic 5G Control: A Theory-Driven Testbed Characterization of Where They Can Be Applied
- National Centre for Scientific Research “Demokritos” (NCSRD)（德谟克利特国家科学研究中心）
机构由 AI 辅助整理，请以论文原文为准。
中文总结 AI 辅助
本文提出理论驱动的类型安全决策框架表征，在5G测试平台上评估三种设计，发现类型安全消除格式错误但微调编码器易误答，而读取问题的框架错误率更低但速度或资源代价高。
中文摘要 AI 辅助
本文提出了一种理论驱动的类型安全决策框架表征方法，用于5G网络的智能体控制，其中每个决策必须是声明选项集合中的元素，而非自由文本。在具有闭环核心策略环路的Open5GS/UERANSIM测试平台上，评估了三个设计点：托管类型模型（Jev）、开放可微调类型编码器（Laya）以及通用语言模型的零标签改造（AnyJev）。所提出的理论框架将及时性、类型一致性、认证成本和基数转化为可检查的适用性谓词，并辅以最优执行/升级/弃权（不执行）门控、升级可行性下限、共置稳定性条件、带认证标签下限的每类型共形风险控制以及类型不匹配界限。对每个谓词进行测量，得到从框架到5G决策类别的适用性映射。类型安全消除了格式错误，但并未消除问题：微调后的类型编码器对98-99.5%的变更问题返回其训练答案，其校准门控随后在多达80%的此类问题上做出错误行动，而读取问题的框架在任意变更问题上最多以0.143（Jev）和0.137（AnyJev）的比例做出错误行动，但前者要么是托管的且慢11-29倍（Jev），要么依赖80亿参数的语言模型（AnyJev）。
英文摘要
This paper presents a theory-driven characterization of type-safe decision frameworks for the agentic control of 5G networks, where every decision must be an element of a declared option set rather than free text. Three design points are evaluated on an Open5GS/UERANSIM testbed with a closed core-policy loop, namely a hosted typed model (Jev), an open fine-tunable typed encoder (Laya), and a zero-label retrofit of a general language model (AnyJev). The proposed theoretical framework turns timeliness, type conformance, certification cost and cardinality into checkable applicability predicates, supported by an optimal act/escalate/abstain gate, an escalation-feasibility floor, a co-location stability condition, per-type conformal risk control with a certification label floor, and a type-mismatch bound. Measuring every predicate yields an applicability map from framework to 5G decision class. Type safety removes format failures but not the question: the fine-tuned typed encoder returned its training answer for 98-99.5% of changed questions, and its calibrated gate then acted wrongly on up to 80% of them, whereas the question-reading frameworks acted wrongly on at most 0.143 (Jev) and 0.137 (AnyJev) of any changed question, but were either hosted and 11-29 times slower (Jev) or reliant on an 8B language model (AnyJev).
