跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2609.11076 · v4

SaltBench: A Referee-Gated Protocol for Measuring Method Effects in Machine-Checked Software Work

SaltBench:一种由裁判门控的协议,用于测量机器可检验软件工作中的方法效应

首次提交 2026-09-10 12:32
版本更新 2026-11-06 00:49
雷达观测 2026-11-06 12:09
arXiv ↗PDF ↗

摘要

SaltBench asks how a machine referee (a proof kernel, program verifier or withheld test suite) changes how a coding agent works. Outcomes are decided outside the agent's toolchain; a wall probed before any scored run isolates the agent from the network, reference solutions and harness; a dated freeze of predictions authorizes each run; a budget stop is a halt, never a failure. Five Rust systems components on a pinned Verus toolchain are each refereed by a withheld test suite. The arms: plain; salt-diet, also instructed to specify and verify its code (a registered reduced rendering of the method); and two arms handed the specification a priori, where the registered sign test at k=4 reached no verdict (3 of 4, p = 0.3125). salt-diet cost more on all five components, by a practical margin: no premium exceeded 2.8879x under either reading of the declared set and the three cheapest sat below 1.4x, a bound of this population and not a promise about larger ones: the premium is near 1 on the smallest components and rises with size. Versions 2 and 3 add the complete pilot matrix: 200 conditions over four models, 181 with a result of record, 16 inexpressible and 3 declared unreached at the cost cap, costed in tokens, dollars and wall time, with no verdict on the arms. Version 3 adds a correctness reading, declared post hoc and registered after every verdict it reads existed and many had been read by its author: 447 cells pass the withheld suite, 40 fail, 10 are censored at a registered budget, 4 are unscorable and 42 are read by condition. salt-diet's pass-rate interval lies wholly below plain's on 8 greenfield and 5 brownfield rows and wholly above it on 2 greenfield rows. It is descriptive, over three cells of record per condition, with no test, and says nothing about whether the method makes code more or less correct. The full record is public.

AI 论文解读

基于摘要 · 非全文审阅

提出 SaltBench 协议,用外部机器裁判(证明内核、程序验证器或保留测试集)来测量编码智能体采用“先规格后验证”方法后的成本与通过率变化。

正在读取账号与研究记录…

版本差异

本站首次收录为 v4,暂无更早版本可对照。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v4 · 2026-11-06 00:49 · 最新观测版本

SaltBench: A Referee-Gated Protocol for Measuring Method Effects in Machine-Checked Software Work

SaltBench asks how a machine referee (a proof kernel, program verifier or withheld test suite) changes how a coding agent works. Outcomes are decided outside the agent's toolchain; a wall probed before any scored run isolates the agent from the network, reference solutions and harness; a dated freeze of predictions authorizes each run; a budget stop is a halt, never a failure. Five Rust systems components on a pinned Verus toolchain are each refereed by a withheld test suite. The arms: plain; salt-diet, also instructed to specify and verify its code (a registered reduced rendering of the method); and two arms handed the specification a priori, where the registered sign test at k=4 reached no verdict (3 of 4, p = 0.3125). salt-diet cost more on all five components, by a practical margin: no premium exceeded 2.8879x under either reading of the declared set and the three cheapest sat below 1.4x, a bound of this population and not a promise about larger ones: the premium is near 1 on the smallest components and rises with size. Versions 2 and 3 add the complete pilot matrix: 200 conditions over four models, 181 with a result of record, 16 inexpressible and 3 declared unreached at the cost cap, costed in tokens, dollars and wall time, with no verdict on the arms. Version 3 adds a correctness reading, declared post hoc and registered after every verdict it reads existed and many had been read by its author: 447 cells pass the withheld suite, 40 fail, 10 are censored at a registered budget, 4 are unscorable and 42 are read by condition. salt-diet's pass-rate interval lies wholly below plain's on 8 greenfield and 5 brownfield rows and wholly above it on 2 greenfield rows. It is descriptive, over three cells of record per condition, with no test, and says nothing about whether the method makes code more or less correct. The full record is public.

查看此版本解读 →查看此版本原文 ↗