# 面向自适应 VLA 动作块执行的策略对齐预测验证

> 原文链接：https://robotics.hellobike.com/papers/2026-10-15-2610-21434
> 频道：papers | 更新日期：2026-10-15 | 标签：arxiv, cs.RO
> 核心摘要：针对VLA动作序列执行长度选择，提出PAV在内部任务条件表示空间评估候选前缀，并在LIBERO、RoboTwin 2.0和真实操纵中验证。

# 面向自适应 VLA 动作块执行的策略对齐预测验证

::: v-pre
> 原文链接：[Policy-Aligned Predictive Verification for Adaptive VLA Action-Chunk Execution](https://arxiv.org/abs/2610.21434)（arXiv 2610.21434 · 发布日期：2026-10-15）

**Policy-Aligned Predictive Verification for Adaptive VLA Action-Chunk Execution**

针对VLA动作序列执行长度选择，提出PAV在内部任务条件表示空间评估候选前缀，并在LIBERO、RoboTwin 2.0和真实操纵中验证。

## 核心要点

- VLA策略生成动作序列可减少重复推理，但执行多少前缀对任务成功至关重要。
- PAV预判候选动作前缀的后果，评估其是否兼容后续成功完成任务。
- 通过轨迹级生存监督和定位先验，从终端任务结果学习局部判断并生成有效性分数。
- 在pi_0和pi_0.5上，跨LIBERO、RoboTwin 2.0及真实操纵改善成功-查询权衡。

**验证范围与局限**：验证范围为pi_0和pi_0.5在LIBERO、RoboTwin 2.0及真实世界操纵任务上的评估。

## 英文摘要 (Abstract)

Vision-language-action (VLA) policies generate action sequences that reduce repeated inference, but deciding how much of each sequence to execute is critical to task success. Existing adaptive methods make this decision using generation stability or prediction agreement, yet these signals do not establish whether executing a candidate prefix preserves the ability to complete the task. We introduce a policy-aligned predictive verifier (PAV) that anticipates the consequences of candidate action prefixes and evaluates their compatibility with successful continuation before execution. Rather than using future prediction to generate actions, our method evaluates existing proposals in the VLA's internal task-conditioned representation space. To learn local judgments from terminal task outcomes, we combine trajectory-level survival supervision with a localization prior that preserves useful behavior within failed rollouts. The resulting validity scores guide execution-horizon selection. Evaluations with pi_0 and pi_0.5 across LIBERO, RoboTwin 2.0, and real-world manipulation show improved success-query trade-offs. These findings point toward a different role for future prediction in robotics, using it to assess how far to trust a policy's proposed actions rather than to generate new ones.

作者：Minkyu Kim, Junhoo Lee, Seungyeon Kim, Suhyun Jeon, Baekseung Kim, Donghoon Han, Nojun Kwak
:::

## 原始参考来源 (Sources)

- [arXiv 2610.21434：Policy-Aligned Predictive Verification for Adaptive VLA Action-Chunk Execution](https://arxiv.org/abs/2610.21434)