PALM：面向长程机器人操作、经由可供性推理的进度感知策略学习
PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulation
把可供性（物体相关性、接触几何、空间摆放、运动动态）作为视觉-运动控制的中间表征来蒸馏，并配合连续子任务进度预测，为 VLA 类模型提供可复用的长程执行稳定化思路。
原文依据 PALM distills complementary affordance representations that capture object relevance, contact geometry, spatial placements, and motion dynamics
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论近期视觉-语言-动作（VLA）模型的进展在机器人操作中展现出潜力，但其在长程、多步任务上仍表现不佳。现有方法缺乏能够识别任务相关交互线索或跟踪子任务内进度的内部推理机制，从而导致重复动作、遗漏步骤和过早终止等关键执行错误。为应对这些挑战，我们提出 PALM，一个围绕以交互为中心的可供性推理与子任务进度线索来组织策略学习的 VLA 框架。PALM 蒸馏出互补的可供性表征，涵盖物体相关性、接触几何、空间摆放与运动动态，并将其作为视觉-运动控制的任务相关锚点。为进一步稳定长程执行，PALM 预测子任务内的连续进度，从而实现无缝的子任务切换。在大量仿真与真实世界实验中，PALM 一致优于基线，在 LIBERO-LONG 上取得 91.8% 的成功率，在 CALVIN ABC->D 上平均完成长度提升 12.5%，并在三个长程泛化设定上相对真实世界基线取得 2 倍提升。
英文原文摘要 1216 字符
Recent advancements in vision-language-action (VLA) models have shown promise in robotic manipulation, yet they continue to struggle with long-horizon, multi-step tasks. Existing methods lack internal reasoning mechanisms that can identify task-relevant interaction cues or track progress within a subtask, leading to critical execution errors such as repeated actions, missed steps, and premature termination. To address these challenges, we introduce PALM, a VLA framework that structures policy learning around interaction-centric affordance reasoning and subtask progress cues. PALM distills complementary affordance representations that capture object relevance, contact geometry, spatial placements, and motion dynamics, and serve as task-relevant anchors for visuomotor control. To further stabilize long-horizon execution, PALM predicts continuous within-subtask progress, enabling seamless subtask transitions. Across extensive simulation and real-world experiments, PALM consistently outperforms baselines, achieving a 91.8% success rate on LIBERO-LONG, a 12.5% improvement in average length on CALVIN ABC->D, and a 2ximprovement over real-world baselines across three long-horizon generalization settings.