arXiv cs.AI· Vasundra Srinivasan·· 14 小时前AI 评分52
研究：computer-use agent 的点击失效多源于页面加载，而非规划间隔
What Does "Now" Mean to Your Computer-Use Agent? Stale Screenshots, Page Loads, and Which Check to Run Before a Click
AI 导读
一项在 31 个公开网页上测量计划点击的研究发现，点击失效的风险主要来自页面加载：在完全加载的页面上规划的点击中，0.4% 在 30 秒内失效，而页面打开后第一秒内规划的点击有 17% 失效。
正文 · 原文
Abstract:Computer-use agents plan on a screenshot for several seconds and then click at a position on the screen; if the page has changed in the meantime, the click lands on an unintended element. Prior work treats the planning interval as the source of this risk. Measuring planned clicks on 31 public web pages, we find that the risk lies in page loading: 0.4% of clicks planned on fully loaded pages became stale within 30 seconds, against 17% of clicks planned in the first second after a page opened. On these pages, waiting about one second for network activity to stop removes most of this risk, yet the wait rules of eight of ten widely used open-source agents leave more than 1% of clicks stale. Commercial pages keep loading for longer: in the HTTP Archive's public crawl, 15% of 65 desktop booking and buying home pages went 500 ms without network requests within 3 seconds of their HTML being parsed, against 63% of 3,028 .gov pages, and 52% shifted their layout within 6.5 seconds of a network-idle wait capped at 3 seconds, against 8% (61% against 26% among pages recorded in full). Where a check before the click is needed, as before irreversible actions and on pages that are still loading, a pixel comparison verifies the target's position as accurately as a frontier vision-language model at one eighth of the latency, and a 4B model matches the frontier model in verifying whether the previous action succeeded. We release a benchmark generator, a break-even formula for when a check pays, and three rules for agent developers: this https URL
| Comments: | 8 pages, 4 figures, 3 tables. Code, benchmark generator and preregistrations: this https URL |
| Subjects: | Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.39282 [cs.AI] |
| (or arXiv:2610.39282v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2610.39282 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Vasundra Srinivasan [view email]
[v1]
Thu, 29 Oct 2026 03:33:26 UTC (1,578 KB)
来源：arXiv cs.AI · arxiv.org