用声明式语言统一编排从推理路由到智能体调度的全流程，自动验证并同步策略变更。
From Inference Routing to Agent Orchestration: Declarative Policy Compilation with Cross-Layer Verification
- 通过非图灵完备的语义路由语言，实现从请求路由到多步智能体工作流的统一编排。
- 单次编译生成可验证的决策节点，覆盖框架、K8s资源及协议边界，支持跨层一致性。
- 阈值修改自动同步至各层级，消除团队间协调导致的策略漂移，适合需要合规与成本控制的场景。
语义路由领域特定语言（Semantic Router DSL）是一种非图灵完备的策略语言，已在生产环境中用于每请求的LLM推理路由：内容信号（嵌入相似度、敏感信息检测、越狱评分）输入加权投影和优先级排序的决策树，以选择模型、执行隐私策略并生成结构化审计日志——全部来自单一声明式源文件。先前工作已建立概率谓词的无冲突编译，并将该语言定位在工作负载-路由器-池的推理架构中。本文将同一语言从无状态的每请求路由扩展至多步骤智能体工作流——涵盖从推理网关到智能体编排再到基础设施部署的完整路径。DSL编译器可输出适用于编排框架（LangGraph、OpenClaw）、Kubernetes资源（NetworkPolicy、Sandbox CRD、ConfigMap）、YANG/NETCONF报文及协议边界门（MCP、A2A）的验证后决策节点——全部源自同一源文件。由于语言为非图灵完备，编译器保证路由全覆盖、分支无冲突、引用完整性及与决策逻辑耦合的审计追踪。因信号定义跨目标共享，阈值变更可在一次编译中从推理网关传播至智能体网关再到基础设施资源，消除跨团队协调作为策略漂移的主要来源。本文基于审计性、成本效率、可验证性和可调性四大支柱，明确了各层的验证边界。
原文摘要 · Abstract (English)
The Semantic Router DSL is a non-Turing-complete policy language deployed in production for per-request LLM inference routing: content signals (embedding similarity, PII detection, jailbreak scoring) feed into weighted projections and priority-ordered decision trees that select a model, enforce privacy policies, and produce structured audit traces -- all from a single declarative source file. Prior work established conflict-free compilation for probabilistic predicates and positioned the DSL within the Workload-Router-Pool inference architecture. This paper extends the same language from stateless, per-request routing to multi-step agent workflows -- the full path from inference gateway to agent orchestration to infrastructure deployment. The DSL compiler emits verified decision nodes for orchestration frameworks (LangGraph, OpenClaw), Kubernetes artifacts (NetworkPolicy, Sandbox CRD, ConfigMap), YANG/NETCONF payloads, and protocol-boundary gates (MCP, A2A) -- all from the same source. Because the language is non-Turing-complete, the compiler guarantees exhaustive routing, conflict-free branching, referential integrity, and audit traces structurally coupled to the decision logic. Because signal definitions are shared across targets, a threshold change propagates from inference gateway to agent gate to infrastructure artifact in one compilation step -- eliminating cross-team coordination as the primary source of policy drift. We ground the approach in four pillars -- auditability, cost efficiency, verifiability, and tunability -- and identify the verification boundary at each layer.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。