外观
论文arXiv 2610.0574723 天前
未过期计划：用于加速 Diffusion Policies 的免费监视器
原文链接：The Unexpired Plan: A Free Monitor for Accelerated Diffusion Policies（arXiv 2610.05747 · 发布日期：2026-10-05）
The Unexpired Plan: A Free Monitor for Accelerated Diffusion Policies
针对无训练扩散策略加速的可靠门控，提出读取未过期计划的监视器，并在114种配置和多个策略家族上验证。
核心要点
- 作者指出，仅用内部相似性信号判定捷径无变化，可能在闭环成功率上出现门控通过但任务结果相反的情况。
- 监视器被拆为读取的统计量和触发后的响应；方法利用分块策略尚未过期的上一计划，衡量相对被替代策略的偏差。
- 在114种加速器配置和四个策略家族上，守护每次调用将单次调用计算量降低1.55--3.09倍。
- 三类策略家族仅靠调度已满足预设±2点裕量；监视器在第四类、未选配置和接触丰富第五类中显示价值。
验证范围与局限：验证范围包括114种加速器配置、四个策略家族及一个接触丰富第五家族；摘要未说明具体机器人或任务。
英文摘要 (Abstract)
Training-free acceleration of a diffusion policy is accepted when an internal similarity signal reports that the shortcut changed nothing. We price every monitor a control loop can afford in closed-loop success rather than feature distance, over $114$ accelerator configurations and four policy families: two accelerators each clear their own gate's bar and log every reuse as certified, while on one task one finishes every episode and the other none. A monitor is two designs, not one --- the statistic it reads, and what it does when that statistic fires. Published gates re-arm after every rejection, and under that response even an oracle handed every forward pass and the exact local action error loses twenty points; absorbing the same statistic costs one point, and most of its speed. What makes a response that never forgets affordable is a statistic that rarely fires, and what it must measure is deviation from the policy the accelerator replaced --- which a chunked policy has already paid for, its last plan not yet expired and free to read. Guarding every call this way cuts per-call compute by $1.55$--$3.09\times$, where any reference-requiring check at the same coverage would have to stop accelerating altogether. On three of our four families the schedule alone already holds the pre-stated $\pm2$-point margin. What the monitor is measurably worth shows in three places: on the fourth family, where it rescues the candidate selection landed on; on four configurations it did not select; and on a contact-rich fifth family, chosen where the schedule was expected to fail and run after every design choice was frozen, where no unmonitored arm at its speed holds the margin and the monitored one does.
作者：Yi Zhao, Sebastian Scherer
来源
本页事实对应的公开页面，以原文为准。