反思式规划：面向多阶段长时程机器人操作的视觉语言模型
Reflective Planning: Vision-Language Models for Multi-Stage Long-Horizon Robotic Manipulation
AI总结：
本文提出一种反思式测试时计算框架，通过生成模型想象未来状态并批判性反思，增强视觉语言模型在多阶段长时程机器人操作中的物理推理能力，显著优于现有商业VLM及MCTS等方法。
摘要：
解决复杂的长时程机器人操作问题需要高级的规划能力、对物理世界的推理能力以及反应性地选择合适的运动技能。在互联网数据上预训练的视觉语言模型（VLMs）原则上可以为解决此类问题提供一个框架。然而，目前形式的VLMs既缺乏机器人操作所需的复杂物理机制的细微理解，也缺乏在长时程上进行推理以解决误差累积问题的能力。在本文中，我们提出了一种新颖的测试时计算框架，以增强VLMs在多阶段操作任务中的物理推理能力。该框架的核心是通过一种“反思”机制迭代地改进预训练的VLM——它使用生成模型来想象未来的世界状态，利用这些预测来指导动作选择，并批判性地反思潜在的次优之处以优化其推理。实验结果表明，我们的方法显著优于多个最先进的商业VLM以及其他后训练方法，如蒙特卡洛树搜索（MCTS）。视频可在https://reflect-vlm.github.io获取。
英文摘要：
Solving complex long-horizon robotic manipulation problems requires sophisticated high-level planning capabilities, the ability to reason about the physical world, and reactively choose appropriate motor skills. Vision-language models (VLMs) pretrained on Internet data could in principle offer a framework for tackling such problems. However, in their current form, VLMs lack both the nuanced understanding of intricate physics required for robotic manipulation and the ability to reason over long horizons to address error compounding issues. In this paper, we introduce a novel test-time computation framework that enhances VLMs' physical reasoning capabilities for multi-stage manipulation tasks. At its core, our approach iteratively improves a pretrained VLM with a "reflection" mechanism - it uses a generative model to imagine future world states, leverages these predictions to guide action selection, and critically reflects on potential suboptimalities to refine its reasoning. Experimental results demonstrate that our method significantly outperforms several state-of-the-art commercial VLMs as well as other post-training approaches such as Monte Carlo Tree Search (MCTS). Videos are available at https://reflect-vlm.github.io.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
