发表机构
Carnegie Mellon University（卡内基梅隆大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出一种基于未过期计划的免费监视器，通过极少触发的偏差统计量守护扩散策略加速中的每次调用，在保持±2分余量的同时将计算量减少1.55至3.09倍，并在多个任务家族中验证其价值。
AI 中文摘要
当内部相似性信号报告捷径未改变任何内容时，扩散策略的无训练加速被视为可接受。我们以闭环成功率而非特征距离来定价每个控制回路能负担的监视器，涵盖114种加速器配置和四个策略家族：两个加速器各自通过自身门槛的基准，并将每次复用记录为已认证，而在一个任务上，一个加速器完成所有回合，另一个则一个未完成。监视器由两个设计组成，而非一个——它读取的统计量，以及当该统计量触发时它采取的行动。已发布的门控在每次拒绝后重新武装，在这种响应下，即使一个拥有每次前向传播和精确局部动作误差的预言机也会损失二十分；吸收相同的统计量仅损失一分，且损失大部分速度。使一个永不遗忘的响应变得可负担的是极少触发的统计量，而它必须衡量的是与加速器所替代策略的偏差——分块策略已为此付出代价，其最后一个计划尚未过期且可免费读取。以这种方式守护每次调用，将每次调用的计算量减少1.55至3.09倍，而在相同覆盖率下任何需要参考的检查将不得不完全停止加速。在我们的四个家族中的三个，仅调度本身就已保持预先声明的±2分余量。监视器的可衡量价值体现在三处：在第四个家族中，它挽救了候选选择所落地的方案；在四个它未选择的配置上；以及在一个接触丰富的第五个家族中，该家族选在调度预期失败且所有设计选择冻结后运行之处，在那里，没有监视的臂在其速度下无法保持余量，而受监视的臂则能保持。
英文摘要
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.
Comments8 pages, 3 figures, 5 tables. Project website (code and videos): https://yizhaojasper.com/unexpired-plan/
