论文/arXiv 2610.05719·23 天前

何时切换：面向视觉-语言-动作模型的可靠动作块扩展

针对VLA长动作块在子技能切换处不可靠的问题，RACE预测切换时机并条件生成动作，在仿真基准和真实机器人上验证。

AI 整理

• 仿真与训练
• 具身模型
• 端侧部署

原文链接：When to Switch: Reliable Action-Chunk Extension for Vision-Language-Action Models（arXiv 2610.05719 · 发布日期：2026-10-05）

When to Switch: Reliable Action-Chunk Extension for Vision-Language-Action Models

针对VLA长动作块在子技能切换处不可靠的问题，RACE预测切换时机并条件生成动作，在仿真基准和真实机器人上验证。

核心要点

• VLA推理开销高会导致策略调用间暂停，stop-and-go执行打断平滑运动并延长任务。
• 长动作块可减少策略调用，但误差集中在操控子技能转换处，并随块长度急剧增长。
• RACE通过辅助一步去噪预测转换时机，并以此条件化动作生成以降低转换误差。
• 仿真中同块长优于微调，2x块长超过近期SOTA与高效VLA，4x仍有竞争力；代码和真机演示已公开。

验证范围与局限：验证范围为仿真基准和一台真实机器人。

英文摘要 (Abstract)

Vision-Language-Action (VLA) models serve as unified policies for robotic manipulation, yet their expensive inference forces robots to pause between policy calls, resulting in stop-and-go execution that interrupts smooth motion and prolongs task completion. Extending the action chunk reduces policy calls and hence these pauses, but predicting farther into the future makes long-chunk execution unreliable. To understand where this unreliability arises, we analyze action errors within long chunks and find that they concentrate around transitions between manipulation subskills, growing sharply with chunk length. This suggests the importance of transition timing, i.e., when to switch subskills within a chunk. Motivated by this observation, we introduce RACE (Reliable Action-Chunk Extension), a framework that predicts the transition timing from an auxiliary one-step denoising pass and conditions action generation on it. By learning and conditioning on transition timing, RACE reduces errors at subskill transitions and enables reliable execution of longer action chunks. Across simulation benchmarks, RACE outperforms fine-tuning at the same chunk length; with 2x longer chunks, it surpasses recent state-of-the-art and efficient VLAs in success rate, and with 4x longer chunks, it remains competitive. On a real robot, RACE uses 4x longer chunks, which reduces the idle time caused by stop-and-go execution by about 5x, while achieving a higher success rate than fine-tuning with the same chunk length. Code and a real-robot demo are available at https://github.com/Seonghoon-Yu/RACE-VLA

作者：Seonghoon Yu, Dongwon Kim, HyungRok Jung, Yoonjae Baek, Byung-kwan Lee, Suha Kwak, Jeany Son

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.05719：When to Switch: Reliable Action-Chunk Extension for Vision-Language-Action Models访问 2026-10-21