REACH：面向扩散策略的显式恢复行为
REACH: Explicit Recovery Behavior for Diffusion Policies
把生成模型中的负提示思想迁移到扩散策略的动作采样上：用误差检测器识别 OOD 执行历史并构造负动作提示，从而在推理时抑制次优动作模态，这一思路可直接复用于视觉-语言-动作模型的动作去噪与采样引导。
原文依据 leverages an error detector to identify out-of-distribution (OOD) execution histories and uses them to construct negative action prompts
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论扩散策略已成为机器人学习的一种强大范式，但其固有的多模态性可能导致在单一观测下产生一组多样且看似合理、却未必最优的动作。我们假设对于给定任务，最优动作存在于该分布之中。受生成模型中负提示的启发，我们提出一种新方法：利用误差检测器识别分布外（OOD）的执行历史，并用这些历史构造负动作提示。由此，我们的策略能够远离次优行为，并收敛到性能更高的动作。我们给出了全面的消融实验，证明正向提示与负向提示的有效性，并在多组仿真基准和真实机器人任务上验证了该方法。结果表明，所提出的负提示引导扩散策略通过有效过滤不良动作模态，在任务性能上取得了显著提升。
英文原文摘要 1008 字符
Diffusion policies have emerged as a powerful paradigm for robot learning, but their inherent multi-modality can lead to a diverse set of plausible--though not always optimal--actions from a single observation. We posit that for a given task, an optimal action exists within this distribution. Inspired by negative prompting in generative models, we introduce a novel method that leverages an error detector to identify out-of-distribution (OOD) execution histories and uses them to construct negative action prompts. This allows our policy to steer away from suboptimal behaviors and converge towards higher-performance actions. We present a comprehensive ablation study demonstrating the effectiveness of positive, and negative prompts, and validate our approach on a suite of simulated benchmarks and real-world robotic tasks. Our results show that the proposed Negative-Prompt-guided Diffusion Policy achieves significant improvement in task performance by effectively filtering undesirable action modes.