让机器人在执行中学会反思，自动优化下一步行动。
Learning Affordances at Inference-Time for Vision-Language-Action Models
- 通过上下文回忆过往经验，动态调整行为策略。
- 在长时序任务中成功识别高可用指令，提升成功率。
- 适合需要持续改进的复杂机器人控制场景。
解决复杂的现实世界控制任务往往需要多次尝试：失败后，我们反思错误并调整策略以避免重复犯错。在机器人领域，视觉-语言-动作模型（VLAs）为解决复杂任务提供了前景，但缺乏在任务失败时动态调整行为的能力。本文提出推理时学习（LITEN），将低层VLA策略与高层视觉语言模型（VLM）连接，通过在上下文中引入过往经验来学习低层策略的可操作性。该方法在推理阶段生成并执行计划，评估阶段则分析实际执行结果（如原始视频轨迹），提炼有效经验并融入后续推理。与非机器人领域的自修正方法不同，LITEN需处理非结构化的现实世界轨迹，因此在评估阶段引入结构化引导。实验表明，LITEN能有效从经验中学习，生成利用高可操作性指令的规划，成功完成长时序任务。
原文摘要 · Abstract (English)
Solving complex real-world control tasks often takes multiple tries: if we fail at first, we reflect on what went wrong, and change our strategy accordingly to avoid making the same mistake. In robotics, Vision-Language-Action models (VLAs) offer a promising path towards solving complex control tasks, but lack the ability to contextually and dynamically readjust behavior when they fail to accomplish a task. In this work, we introduce Learning from Inference-Time Execution (LITEN), which connects a VLA low-level policy to a high-level VLM that conditions on past experiences by including them in-context, allowing it to learn the affordances and capabilities of the low-level VLA. Our approach iterates between a reasoning phase that generates and executes plans for the low-level VLA, and an assessment phase that reflects on the resulting execution and draws useful conclusions to be included in future reasoning contexts. Unlike similar approaches to self-refinement in non-robotics domains, LITEN must reflect on unstructured real-world robot trajectories (e.g., raw videos), which requires structured guiderails during assessment. Our experimental results demonstrate LITEN is able to effectively learn from past experience to generate plans that use high-affordance instructions to accomplish long-horizon 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。