跳到正文
研踪PaperTrail

论文库

检索计算机领域论文,按研究方向、作者或关键词筛选。

部分采集异常
正在读取账号与研究记录…

全站更新概况

+1112新增论文
+875其中 AI 相关
+133已有论文修订
+94新增 AI 导读

arXiv 最近完成扫描:2026-11-19 23:27 · 全站最近入库:2026-11-19 19:09各信源进度 →

统计口径与数据新鲜度

统计 2026-11-18 23:30 — 2026-11-19 23:30(北京时间) · 点击数字查看对应简报;AI 相关包含在新增论文内。

最近完成检查:2026-11-19 23:27最近发现新增版本:2026-11-19 19:09库内论文最新更新时间:2026-11-19 02:59

每分钟刷新已保存的数据。空检查不代表采集异常;新收录可能包含历史论文。AI 导读按新增版本计数。

27,982 已收录论文23,249 AI 相关8,559 有修订112 HF 社区发现

浏览论文库

时间范围跟随排序:版本更新、本站收录或首次发表。最新收录可能包含历史论文。部分信源异常,查看状态 →

找到 1 篇论文 · 按版本更新时间排序上轮扫描起点 2026-11-19 23:25(北京时间)
版本更新 · v2arXiv:2609.32495

Hearsay: Who Can Audit the Record a Deployed Agent Harness Writes?

Jiahong Dai, Zhuochen Yang, Pengyang Shao, Kelvin Ng, Zhongyi Liu, Chengquan Ju, Yuting He, Bo Hu

原文摘要

The record of an agent's run is written by the harness that drives the agent. When the run is later disputed, the auditor was not there and has only this record. We ask whether someone who was not there can check the record, and what decides the answer. It depends on who wrote what a check can rely on: if the harness wrote all of it, the record is only the harness's own word, and we call it hearsay. We build Hearsay, a benchmark that examines the record and not the task: five harnesses run fourteen tasks, model examiners from three vendors and a human panel read the records and answer questions about the runs, and every passage they cite is labelled mechanically by who wrote it. We also audit what is deployed: in all sixteen frameworks, the harness is the only writer. With one writer, readers can name the fault from the record, in up to nine runs in ten, but they cannot check the record. The model examiners claim corroboration for 18 to 55 points more of their evidence than the labels support, a gap we then predicted for runs driven by another vendor's model and found again. And the best of them notices only half of the alterations we make to a record, while the two that name the fault most often cannot tell what a harness left out. With a second writer, the record can be checked. Our sandbox keeps a ledger of what passes between harness and model, and reading the ledger against the record in both directions shows what the record leaves out and what it adds. We test this as an auditor would use it, on one unseen record at a time. On runs driven by a third vendor's model, with every rule committed before the records existed, it reports 64 of 66 alterations and flags none of 84 clean records. On held-out records in five formats it reports 109 of 115. What makes a record evidence is a second writer, not a better reader.

cs.CR · cs.AI · cs.CE · cs.SE