治理审查自动化替代人类把关的研究框架与 DGF-Bench 评测
The Last Human Gate: Forward Deployed Engineering for Governance Automation
几个新模型在治理审查门上分数差距很大，完整路线跑通率最高也只有七成六，作者把为什么开源了，做流程自动化的可以看看。
该论文提出任务替代框架，将数字治理框架（DGF）中的每个审查门视为可执行契约，并推导残余工作量阈值，说明自动化多数案例后总人力反而可能上升。基于 300 个合成项目和 899 次模型-项目运行，Gemini 3.8 Flash、GPT-5.6 Luna、DeepSeek v4.1 Flash 的严格门通过率分别为 94.98%、83.29% 和 74.18%，完整路线通过率则为 76.92%、42.33% 和 24.67%。确定性控制程序在给定规则与结构化事实下通过全部 1,700 个门，说明瓶颈在执行而非决策。研究通过证据审计和 135 次重复运行区分正确决策与可靠执行，源码与数据均已公开。
The Last Human Gate: Forward Deployed Engineering for Governance Automation
Enterprise governance requires decisions, evidence, and accountable authority; it does not require every review task to retain its current human implementation. We develop a task-substitution framework for Digital Governance Frameworks (DGF), treating each gate as an executable contract. Substitution requires sufficient accessible information, valid decision and authority checks, and a reduction in total human work after exceptions, verification, correction, and maintenance are counted. We derive a residual-work threshold and show why automating most cases can still increase labor. Forward deployed engineering connects these conditions to an architecture for agents, rule engines, evidence services, and escalation. DGF-Bench supplies controlled evidence from 300 synthetic projects and 899 evaluable model-project runs. Gemini 3.8 Flash, GPT-5.6 Luna, and DeepSeek v4.1 Flash achieve strict gate success of 94.98%, 83.29%, and 74.18%; complete-route success is 76.92%, 42.33%, and 24.67%. A deterministic control passes all 1,700 gates given the supplied rules and structured facts, locating the comparison in execution of a supplied decision kernel. Evidence audits and 135 repeated runs distinguish correct decisions from reliable execution. A document counterexample establishes an information-sufficiency obstruction. These results support the technical feasibility of replacing human execution of specified governance-review tasks with agents and software. The framework specifies a workforce test based on the complete human effort required at fixed output and quality; the present measurements concern review performance. Sources, dossiers, traces, and analyses are public.