跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.21407 · v1

Harness Leverage Across Model Size: A Ladder of Five Local Agent Harnesses from 0.75B to 4.7B

首次提交 2026-11-17 01:00
版本更新 2026-11-17 01:00
雷达观测 2026-11-17 16:10
arXiv ↗PDF ↗

摘要

A harness turns a model into a system that completes tasks: it builds the prompt, exposes tools and decides when work is done. Harness comparisons hold the model fixed. Model comparisons hold the harness fixed. We vary both: five shipped harnesses, 18 artifact-producing tasks, four locally deployable open-weight models from 0.75B to 4.7B and a 35.5B anchor, with one deterministic checker scoring all 360 ladder cells. The completion-controlling harness, Mingbird, has the highest mean at every size. Of four pre-registered paired contrasts, two survive Holm correction, at 0.75B and 4.7B. Thirteen of sixteen paired intervals exclude zero under the deployment reading, with leads from 0.075 over the strongest rival to 0.819 over the weakest. The completion gate is not the carrier: disabling all six executable checks moves the mean by +0.080, +0.117, +0.003 and -0.004, every interval including zero, and only at 1.9B does disabling it create empty runs. Removing one repeat-blocking rule alone costs 0.411 at 0.75B and leaves eight of eighteen tasks with no artifact. The harness effect is a progress effect: it decides whether a run leaves anything on disk. Of 357 usable cells, 127 leave none of their declared artifacts and 85 of those exit normally, so a scorer that reads exit status credits work that was never written. Whether the effect grows or shrinks with size, four rungs cannot settle: a floor arm produces a rise, a ceiling a decline, and a cross-family rung a peak.

作者备注:56 pages, 12 tables, 5 figures

AI 论文解读

基于摘要 · 非全文审阅

此版本尚未生成 AI 解读。解读由后台任务生成并按论文版本保存,登录后共享已有结果。

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

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。

v1 · 2026-11-17 01:00 · 最新观测版本

Harness Leverage Across Model Size: A Ladder of Five Local Agent Harnesses from 0.75B to 4.7B

A harness turns a model into a system that completes tasks: it builds the prompt, exposes tools and decides when work is done. Harness comparisons hold the model fixed. Model comparisons hold the harness fixed. We vary both: five shipped harnesses, 18 artifact-producing tasks, four locally deployable open-weight models from 0.75B to 4.7B and a 35.5B anchor, with one deterministic checker scoring all 360 ladder cells. The completion-controlling harness, Mingbird, has the highest mean at every size. Of four pre-registered paired contrasts, two survive Holm correction, at 0.75B and 4.7B. Thirteen of sixteen paired intervals exclude zero under the deployment reading, with leads from 0.075 over the strongest rival to 0.819 over the weakest. The completion gate is not the carrier: disabling all six executable checks moves the mean by +0.080, +0.117, +0.003 and -0.004, every interval including zero, and only at 1.9B does disabling it create empty runs. Removing one repeat-blocking rule alone costs 0.411 at 0.75B and leaves eight of eighteen tasks with no artifact. The harness effect is a progress effect: it decides whether a run leaves anything on disk. Of 357 usable cells, 127 leave none of their declared artifacts and 85 of those exit normally, so a scorer that reads exit status credits work that was never written. Whether the effect grows or shrinks with size, four rungs cannot settle: a floor arm produces a rise, a ceiling a decline, and a cross-family rung a peak.

查看此版本解读 →查看此版本原文 ↗