面向人机协作中具身代理的实时适应
Towards Real-time Adaptation of Embodied Agent in Human-Robot Collaboration
AI总结：
本文提出MonTA框架，通过轻量级监控和高效适配器提升人机协作中具身代理的实时适应能力，实现更高效的协作性能。
摘要：
大型语言模型（LLMs）为人类-机器人协作开辟了变革性的可能性。然而，实现实时协作需要低延迟和鲁棒推理，而大多数LLMs存在高延迟的问题。为解决这一差距，我们首先提出一个细粒度基准，专门评估代理在Overcooked-AI环境中的主动适应性和时间响应性。基于评估结果，我们提出MonTA（监控-然后适应），一个受认知科学研究启发的分层框架。MonTA包含三个关键模块：一个轻量级的监控器，以高频（7 Hz）运行以检测适应需求，以及两个高效的适配器，用于子任务和路径适应推理，以较低频率向人类提供指令。我们的结果表明，MonTA在我们提出的基准上显著优于基线代理，在不同布局中均表现出优越的性能。用户研究确认了我们的框架向人类提供的适应计划的高合理性以及一致的语言指令。
英文摘要：
Large Language Models (LLMs) have opened transformative possibilities for human-robot collaboration. However, enabling real-time collaboration requires both low latency and robust reasoning, and most LLMs suffer from high latency. To address this gap, we first propose a fine-grained benchmark that explicitly assesses agents' proactive adaptability and temporal responsiveness in the Overcooked-AI environment. Based on evaluation results, we propose MonTA (Monitor-then-Adapt), a hierarchical framework inspired by cognitive science research. MonTA contains three key modules: a lightweight Monitor that operates at high frequency (7 Hz) to detect adaptation needs, and two proficient Adapters for subtask and path adaptation reasoning that provide instructions to humans at a lower frequency. Our results demonstrate that MonTA significantly outperforms baseline agents on our proposed benchmark, achieving superior performance across layouts with varying teaming fluency. User studies confirm the high reasonableness of adaptation plans and consistent language instructions provided by our framework to humans.
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。
