发表机构
Independent Researcher Bentonville, AR USA; Independent Researcher Herndon, VA USA; Independent Researcher Bothell, WA USA; Independent Researcher（; ; ; ）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
KNDB 是一种 PostgreSQL 表访问方法，通过引擎分配的认知类型解决写入冲突，在对抗性置信度伪造下显著优于基线，且其优势不可伪造。
AI 中文摘要
我们描述了 KNDB，一种 PostgreSQL 18 表访问方法（TAM），它为每一行赋予引擎分配的认知类型（MEASURED、INFERRED 或 DERIVED），并在每次写入时的 heapam 回调中解决每个槽位的冲突。行作为普通堆元组存储；44 个 TAM 回调中有 7 个被覆盖（tuple_insert、multi_insert、tuple_update、tuple_delete、tuple_insert_speculative、tuple_complete_speculative、relation_toast_am），其余 37 个委托给堆；我们提供了关于该接口的完备性论证作为论文工件。本文报告了该决策背后的工程实现以及促使该决策的对抗性评估。在置信度伪造工作负载中，攻击者以 [0.95,1.0] 的置信度断言 INFERRED 写入，而诚实 MEASURED 写入的置信度为 [0.5,0.9]，KNDB 在 Book-Author 融合数据集上比仅基于置信度的基线高出 63 个百分点，在 Zheng 众包数据集上高出 92.7 个百分点。这两项胜利在类型轴上被证明是承重的，通过源重建禁用测试（其中格被中和，胜利消失）得到验证。针对四个真值发现基线（TruthFinder、CRH、CATD、ACCU），这些基线根据原始方程重新实现并验证到与已发表数字相差 0.3 个百分点以内，KNDB 在每数据集密度饱和单元以下具有竞争力，在饱和单元及以上则占主导地位。我们将该单元形式化为 k* ~ rho_alg * h_top，其中 h_top 是每槽位顶部诚实表面形式支持，并在 Book-Author 上验证预测在 +/-20% 以内，在 Zheng 上在 +/-30% 以内。由于类型轴由引擎根据独立元数据分配，无法在写入时伪造，因此 KNDB 的 k* 是无界的。本文诚实说明了 KNDB 的不足之处：在 Zheng 上低于饱和时 CRH 和 ACCU 优于 KNDB，并且 KNDB 在三个时间知识编辑基准上得分为零，这些基准的真实标签是最后写入者获胜。
英文摘要
We describe KNDB, a PostgreSQL 18 table access method (TAM) that types every row with an engine-assigned epistemic kind (MEASURED, INFERRED, or DERIVED) and resolves per-slot conflicts inside every write-time heapam callback. Rows land as ordinary heap tuples; seven of the 44 TAM callbacks are overridden (tuple_insert, multi_insert, tuple_update, tuple_delete, tuple_insert_speculative, tuple_complete_speculative, relation_toast_am), the other 37 delegate to heap; we provide a completeness argument over the interface as a paper artefact. This paper reports the engineering behind that decision and the adversarial evaluation that motivated it. On a confidence-forgery workload where an attacker asserts INFERRED writes with confidence in [0.95,1.0] against honest MEASURED writes with confidence in [0.5,0.9], KNDB beats a confidence-only baseline by 63 percentage points on the Book-Author fusion dataset and 92.7 points on the Zheng crowdsourcing dataset. Both wins are proven load-bearing on the kind axis by a source-rebuild disable-and-test in which the lattice is neutralised and the win vanishes. Against four truth-discovery baselines (TruthFinder, CRH, CATD, ACCU) reimplemented from the original equations and validated to within 0.3 percentage points of the published numbers, KNDB is competitive below a per-dataset density-saturation cell and dominant at or above it. We formalise the cell as k* ~ rho_alg * h_top, where h_top is per-slot top honest surface-form support, and validate the prediction within +/-20% on Book-Author and +/-30% on Zheng. Because the kind axis is assigned by the engine from independent metadata and cannot be forged at write time, KNDB's k* is unbounded. The paper is honest about where KNDB loses: CRH and ACCU outperform KNDB below saturation on Zheng, and KNDB scores zero on three temporal knowledge-editing benchmarks whose ground truth is last-writer-wins.
Comments13 pages, 1 figure. Under review at PVLDB Volume 20. Code and benchmark artifacts: https://github.com/emailvenkatm/kndb (branch postgres-experiment)
