外观
论文arXiv 2610.1774615 天前
人类--VLA 通信鸿沟：人类留下了什么隐含信息以及策略无法恢复什么
原文链接：The Human--VLA Communication Gap: What Humans Leave Implicit and What Policies Cannot Recover（arXiv 2610.17746 · 发布日期：2026-10-13）
The Human--VLA Communication Gap: What Humans Leave Implicit and What Policies Cannot Recover
研究人类自然指令与VLA策略间的通信鸿沟，通过信息与执行缺口划分，在受控物理任务变体上验证。
核心要点
- 动机是机器人会执行任务，却可能因人类自然请求中的隐含信息而失败。
- 方法将鸿沟分为语义、落地或事件结构未解的信息缺口，以及剩余执行缺口。
- 实验详细分析π_{0.5}，并在π_0-FAST和Cosmos3-Edge上观察到一致趋势。
- 项目网站已提供；发现当前策略难以重建任务语义或事件结构。
验证范围与局限：验证范围为受控的相同物理任务变体，详细分析集中在π_{0.5}。
英文摘要 (Abstract)
A robot may know how to perform a task yet fail when a human asks for it naturally. We study this human--VLA communication gap by separating an \emph{information gap}, where task semantics, grounding, or event structure remain unresolved, from a residual \emph{execution gap}, where even a fully specified task depends on how it is organized and invoked. To do this, we characterize how people naturally express demonstrated objectives and evaluate controlled variants of the same physical tasks, with detailed analysis on $π_{0.5}$ and consistent trends on $π_0$-FAST and Cosmos3-Edge. Our study reveals that humans increasingly compress instructions as tasks grow, relying more on contextual grounding and implicit execution structure. Current policies are misaligned with this behavior: they tolerate grounding dependence relatively well, but struggle when task semantics or event structure must be reconstructed. Even after the information gap is resolved, execution remains sensitive to command granularity and invocation state. These findings identify concrete targets for human-centered policy learning and interfaces that better bridge natural human objectives to robot behavior. Project website: https://commgap.github.io/.
作者：Yiqing Xu, Jiajun Wu
来源
本页事实对应的公开页面，以原文为准。