arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2510.12072cs.AIcs.RO
收藏

EmboMatrix:一种可扩展的具身决策训练场

EmboMatrix: A Scalable Training-Ground for Embodied Decision-Making

Shanghai Jiao Tong University(上海交通大学);Zhongguancun Academy(中关村学院);New York University(纽约大学)

Zixing Lei, Sheng Yin, Yichen Xiong, Yuanzhuo Ding, Wenhao Huang, Yuxi Wei, Qingyao Xu, Yiming Li, Weixin Li, Yunhong Wang, Siheng Chen

AI总结:

提出首个具身决策训练场EmboMatrix,通过多智能体数据引擎、异构硬件模拟和多级奖励架构,使7B模型EmboBrain超越671B基线9.5%,验证交互式环境学习的力量。

摘要:

具身决策使智能体能够通过与物理世界的持续交互,将高层目标转化为可执行的动作,这构成了通用具身智能的基石。大型语言模型(LLMs)凭借其通用决策能力,为实现这一潜力提供了有希望的途径;然而,仅基于语言训练的LLMs缺乏对物理环境的接触,限制了其真正的具身理解。为弥合这一差距,我们提出了训练场(training ground)的概念:一种综合基础设施,提供任务与场景模拟、具身交互和反馈信号,为LLM获得真正的具身决策技能提供一站式解决方案。在本工作中,我们提出了EmboMatrix,这是首个此类训练场,提供海量多样的任务以及高效模拟和精确奖励。EmboMatrix融合了一系列新颖技术:用于大规模任务和场景生成的多智能体数据引擎、用于可扩展模拟的分布式异构硬件系统,以及用于精确监督的多级奖励架构。利用EmboMatrix,我们培育了EmboBrain,一个其具身决策能力从广泛具身交互中涌现的LLM。实验表明,EmboBrain-7B在两个具有挑战性的具身决策基准上,比671B的DeepSeek-R1基线高出9.5%,展示了交互式、环境接地学习在构建真正智能具身智能体方面的力量。

英文摘要:

Embodied decision-making enables agents to translate high-level goals into executable actions through continuous interactions within the physical world, forming a cornerstone of general-purpose embodied intelligence. Large language models (LLMs), with their general decision-making capabilities, offer a promising path to realize this potential; however, LLMs trained solely on language lack exposure to physical environments, limiting their true embodied understanding. To bridge this gap, we propose the concept of a training ground: a comprehensive infrastructure that provides task and scene simulation, embodied interaction, and feedback signals, offering a one-stop solution for LLM acquire genuine embodied decision-making skills. In this work, we present EmboMatrix, the first training ground of its kind, providing massive and diverse tasks with efficient simulation and precise rewards. EmboMatrix incorporates a series of novel techniques: a multi-agent data engine for large-scale task and scene generation, a distributed heterogeneous-hardware system for scalable simulation, and a multi-level reward architecture for precise supervision. Leveraging EmboMatrix, we cultivate EmboBrain, an LLM whose embodied decision-making abilities emerge from extensive embodied interactions. Experiments show that EmboBrain-7B surpasses the 671B DeepSeek-R1 baseline by 9.5\% on two challenging embodied decision-making benchmarks, demonstrating the power of interactive, environment-grounded learning for building truly intelligent embodied agents.

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。

↑