When to Ask for Help: A Decision-Theoretic Account and Measurement Standard for Slow-Module Scheduling in Dual-System Vision-Language-Action Models
作者: Shilin Zhang
分类: cs.RO
发布日期: 2026-10-15
备注: 41 pages, 7 figures. Preprint
💡 一句话要点
提出决策理论框架以优化双系统视觉语言行动模型中的慢模块调度
🎯 匹配领域: 支柱九:具身大模型 (Embodied Foundation Models)
关键词: 双系统模型 视觉语言行动 决策理论 慢模块调度 价值干预 闭环测试 多模态学习
📋 核心要点
- 现有的双系统VLA模型在慢模块调度上缺乏统一的目标和标准,导致调度质量与模块质量难以区分。
- 论文提出将慢模块调用形式化为经过校准的决策问题,利用价值干预(VoI)来优化调度策略。
- 实验结果表明,适应性触发器的查询分配与理论预测一致,且在LIBERO测试中反转执行器表现出零净负面影响。
📝 摘要(中文)
双系统视觉语言行动(VLA)模型结合了高频快速策略与慢速昂贵模块,现有触发机制缺乏共享目标、成本测量及调度质量与模块质量的标准。本文将调用形式化为一个经过校准的决策问题,提出了固定比例、固定预算和历史门控调度的概念,并证明在状态独立崩溃下,最佳调度是周期性的。通过对不同成本比率的系统进行测量,验证了低成本比率下的预测,并在LIBERO上测试了闭环臂,结果显示适应性触发器的查询分配符合阈值规则预测,但语言重写通道未能成功转化查询,反转执行器是唯一在两个测试中无负面影响的干预手段。最后,提出的协议显著降低了早期预警的AUC值。
🔬 方法详解
问题定义:本文旨在解决双系统VLA模型中慢模块调度的缺乏标准化和目标的问题,现有方法未能有效区分调度质量与模块质量。
核心思路:通过将慢模块调用视为一个经过校准的决策问题,提出了基于价值干预(VoI)的调度策略,旨在优化调用时机和频率。
技术框架:整体架构包括三个主要部分:1) 调用机制的形式化;2) 通过测量不同成本比率的系统进行验证;3) 在LIBERO上进行闭环测试以评估调度效果。
关键创新:最重要的创新在于将调度问题转化为决策理论框架,提供了一个量化的标准来评估慢模块的调用时机和效果,这与现有方法的定性分析形成鲜明对比。
关键设计:关键参数包括成本比率$ρ$的设置,采用的损失函数和调度策略的设计,确保在不同环境下的适应性和有效性。
🖼️ 关键图片
📊 实验亮点
实验结果显示,采用的四测漏协议将早期预警的AUC值从0.93降低至0.62,验证了检测质量与调度能力之间的关系。此外,测量的价值函数在每个任务的刷新周期内与闭环oracle的偏差仅为2.4%。
🎯 应用场景
该研究的潜在应用领域包括智能机器人、自动驾驶系统和人机交互等场景,能够有效提升系统在复杂环境中的决策能力和响应效率。未来,该方法有望在多模态学习和实时决策系统中发挥重要作用。
📄 摘要(原文)
Dual-system vision-language-action (VLA) models pair a high-frequency fast policy with a slow, expensive module, and published trigger mechanisms for invoking it lack a shared objective, measured costs, and any standard separating schedule quality from module quality. We provide the missing account: invocation is formalized as a calibrated decision problem whose fixed-ratio, fixed-budget, and history-gated schedules are degenerate cases of one quantity, the Value of Intervention (VoI); a renewal baseline proves the optimal schedule periodic under state-independent collapse, with adaptive gains governed by the cost ratio $ρ$ and the dispersion of collapse. Measured systems span three orders of magnitude in $ρ$ (FiS-VLA $0.95$; RoboICL about $10^3$), and a direct test confirms the low-$ρ$ prediction: calibrated gains vanish under homogeneous collapse and grow with heterogeneity. Fourteen closed-loop arms on LIBERO test the contested middle of the continuum ($ρ=1.53$): the adaptive trigger allocates its queries exactly as the threshold rule predicts, yet the language-rewrite channel cannot convert them into success on either suite, and a revert actuator that undoes a failed rewrite is the only intervention with zero net downside on both LIBERO-Spatial and LIBERO-Goal, where the rewrite deficit deepens from thirteen to forty-two points. A four-measure leak-free protocol cuts apparent early-warning AUC from $0.93$ to $0.62$ on $929$ deployment traces, a detection-limited frontier theorem converts detection quality into a schedulability statement, and a compilation loop turns measured value functions into per-task refresh periods within $2.4\%$ of the closed-loop oracle. Together these results turn slow-module invocation from a design opinion into a measured decision.