动态调整机器人执行节奏，让动作模型更聪明地决定何时重规划。
Rethink Before You Execute: Adaptive Execution for World Action Models
- 通过在线监测任务进展，动态决定是否继续执行当前动作块
- 真实机器人上减少26.9%推理次数，难任务成功率提升13.3点
- 轻量级插件设计，适配各类世界动作模型
世界动作模型（WAMs）联合预测未来动作与环境演化。推理时，WAM生成一段动作并固定执行前缀后重新规划。我们指出这种固定执行窗口与实际执行动态不匹配：动作块可靠性随任务阶段变化，重规划时机应由累积执行结果决定，而非步骤数。为此提出TempoWAM（基于进度监测的动态执行），其包含一个递归进度监测器，从当前观测、任务指令、剩余动作和执行历史中估计任务进展；再由自适应执行协议评估动作块是否推进任务，决定是否重规划。为弥合训练-部署差距，协议通过任务相关校准因子实现在线自适应。在LIBERO、RoboTwin及真实任务上的实验表明，TempoWAM持续提升WAM执行的效率-成功率权衡。在真实机器人上，简单任务减少26.9%推理次数且保持成功，复杂任务成功率提升13.3个百分点。
原文摘要 · Abstract (English)
World Action Models (WAMs) jointly predict future actions and the evolution of the environment. At each inference, a WAM generates a chunk of actions and the robot executes a fixed prefix before replanning. We argue that this fixed execution horizon is poorly matched to execution dynamics: the chunk reliability varies across task stages, so when to replan depends on the result of accumulated execution, not on the step counts. We propose TempoWAM (Timing Execution by Monitoring Progress Online), a lightweight plug-and-play execution scheme for WAMs. A Recurrent Progress Monitor first estimates task progress from the current observation, task instruction, remaining actions, and execution history; and an Adaptive Execution Protocol then evaluates whether the chunk is advancing the task to decide if replanning is needed. To bridge the training-deployment gap, the protocol is calibrated by a task-dependent calibration factor with online adaptation. Experiments on LIBERO, RoboTwin, and real-world tasks show that TempoWAM consistently improves the efficiency-success trade-off of WAM execution. On real robots, it reduces WAM inferences by 26.9% on easy tasks while maintaining success, and improves success by 13.3 points on difficult tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。