---
title: "Quipu：面向智能体的受治理双时序知识图谱存储，将审计化为查询"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.16813v1"
permalink: "https://feed.astralor.com/items/itm_01f0e8bef73bc4e97e901f32"
published: "2026-08-17T17:04:29.000Z"
score: 77
channel: "ai"
tags: ["Agent治理","arXiv","Quipu","双时序","知识图谱"]
---

# Quipu：面向智能体的受治理双时序知识图谱存储，将审计化为查询

## AI 点评

> Quipu 的价值不在于又多一个图存储，而在于把“治理”从外围仪表盘挪进存储内核，让审计变成查询。对做 Agent 基础设施、知识图谱和合规系统的人，这篇值得精读；如果你想给多智能体系统加“事实可信度”和“事后可追责”，它的格与签名裁决设计可以直接借鉴。

## 速览

论文提出Quipu，一种面向智能体知识图谱写入的可嵌入存储，反转四项默认设置：写入前先通过治理门禁、数据/信任/裁决/规则全比特emporal、以命名图为权威与信任单位并保证组合不扩大权限、治理规范与审计轨迹本身也作为图内可查询事实。在Census与DEMM-Bench评测中，门禁消除植入缺陷，审计可复现，并发现真实执行缺口。

## 关键点

- Quipu 将写入门禁、比特emporal 数据模型、命名图信任格与“治理即图内查询”结合，使审计变成 T⊨Σ。
- 在 Census 多写者生命周期中，门禁存储 0/6 植入缺陷，无门禁 6/6；7 项组合探针全部满足格约束。
- 50/50 满意裁决按时间点重推导一致，而“仅取最新规则”全部错报；SARC 审核与店内审计逐裁决一致。
- 在 DEMM-Bench 外部基准上，内容级读取在 8 种降级条件下 512 个问题零过度断言，并暴露并修复否决裁决的证明缺口。

## 全文

Agents now write knowledge graphs, but knowledge-graph stores still carry defaults set when humans curated them: accept writes now and clean later, keep one time axis or none, treat every writer's facts as equally trustworthy, and leave governance to dashboards and middleware. These four defaults are individually convenient and jointly untenable under agent workloads. We present Quipu, an embeddable store that inverts all four: no fact enters except through a gate whose predicates evaluate the pending post-state; data, trust labels, verdicts, and the rules themselves are bitemporal; named graphs are the unit of authority and trust, composed under a lattice whose one invariant is that composition never widens; and the governance specification $Σ$, the trace, and signed verdicts are facts in the store they govern, making the audit $T \models Σ$ a query. We evaluate with Census, a deterministic multi-writer lifecycle whose single seeded run scores every research question against planted ground truth: the gated store ends with 0 of 6 planted defects versus 6 of 6 ungated; all 7 composition probes uphold the lattice contract; 50 of 50 satisfied verdicts re-derive faithfully as of their instant while all 50 would be misreported under a latest-only rule set; and the SARC reference checker agrees with the in-store audit verdict-for-verdict, differing only on coverage semantics. A recorded trace from a governed writer surfaces a live enforcement gap the audit names with its remediation. On DEMM-Bench, an external decision-evidence sufficiency benchmark, a content-only reading of the exported records answers all 512 property-level governance questions correctly with zero overclaim under all eight degradation conditions, while container-presence baselines overclaim on up to 87.5% of them -- and the run surfaced, and led us to close, a gap in what a denial's verdict attests.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_01f0e8bef73bc4e97e901f32