KORA：面向 AI 工作负载的确定性优先执行控制
原标题：KORA: Deterministic-First Execution Control for AI Workloads
AI 摘要
KORA 提出一种「确定性优先」的执行控制层，在推理之前将请求表示为任务路径，优先尝试确定性或结构化路由并验证结果，仅在确定性处理不足时才升级到模型推理。论文报告在一个可复现的 100 项确定性偏重基准中，相比朴素直接基线避免了 80 次模拟模型调用，且确定性子集零不匹配；另有一个合成客服本地/无网络验证工作负载。作者明确限定这些结果不证明生产行为、真实供应商结果、能耗或 API 成本降低。
正文节选
KORA: Deterministic-First Execution Control for AI Workloads Abstract AI applications often invoke a model before they determine whether inference is necessary. That default is useful for ambiguous generative tasks, but it is poorly matched to work that is structured, deterministic, policy-bound, lookup-oriented, or directly verifiable. KORA introduces deterministic-first execution control: a pre-inference control layer that represents requests as task paths, attempts deterministic or structured