发表机构
Incredibuild; HIT CS Department（Incredibuild公司; 以色列理工学院计算机科学系）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
研究在局部渐近正态性下工作节点发出的置信密度，指出其为吉布斯 - 玻尔兹曼测度，逆温度是样本大小。高斯/线性下相关结果精确，MapReduce归约是配分函数，还有频率主义一致性等结论。
AI 中文摘要
在局部渐近正态性（LAN）下到主导阶，工作节点在大小为\(n\)的块上发出的置信密度是一个吉布斯 - 玻尔兹曼测度\(\exp\{-\beta E(\theta)\}\)，其逆温度为样本大小\(\beta = n\)。在高斯/线性情况下这三个结果是精确的，否则是一阶的：不相交块携带独立的玻尔兹曼因子，所以MapReduce的归约实际上是一个配分函数\(Z=\int\prod_k h_k\,d\theta\)，其模式是精度加权（逆方差）合并；频率主义一致性是零温度极限\(T = 1/n\to0\)。
英文摘要
Snapshot-backed sandboxes make branching cheap while leaving evidence dependence unchanged. Branches can reuse a model, prompt, repository, tests, observations, or execution ancestor, so counting outputs can amplify one repeated error into high-confidence consensus. We introduce an \emph{evidence-aware reduction contract}: each worker reports an estimate, estimated information, evidence identifiers, fork lineage, and execution metadata. For independent workers estimating one common parameter, we use standard inverse-information pooling in its Gaussian/Wald form. The fixed-dimensional numeric summary can merge in any tree order; evidence IDs and lineage follow separate rules. The residual $Δ$ measures disagreement, becomes Cochran's $Q$ in the scalar inverse-variance case, and appears in the product integral. A reference implementation validates serialized records, rejects repeated nonempty evidence identifiers, carries evidence and lineage through tree reduction, and uses Cholesky-based numerical linear algebra. Unit tests and seeded synthetic checks exercise the algebra, unequal information, and forged precision; one four-worker named-snapshot trace exercises the end-to-end path. Platform logs document the exercised execution paths. A central open systems challenge is to turn evidence identity and fork lineage into a dependence model for correlated and adaptively selected AI branches.
CommentsN/A
