---
title: "Being-0: A Humanoid Robotic Agent with Vision-Language Models and Modular Skills"
canonical_url: "https://www.modelscope.cn/papers/127565"
md_url: "https://www.modelscope.cn/papers/127565.md"
arxiv_id: 2503.12533
published: 2025-05-11
last_updated: 2025-05-11
authors:
  - "Haoqi Yuan"
  - "Yu Bai"
  - "Yuhui Fu"
  - "Bohan Zhou"
  - "Yicheng Feng"
  - "Xinrun Xu"
  - "Yi Zhan"
  - "Börje F. Karlsson"
  - "Zongqing Lu"
model_name: Being-0
model_developer: "北京大学、北京智源人工智能研究院、BeingBeyond"
domain:
  - "机器人学"
  - "具身智能"
  - "视觉语言模型"
  - "强化学习"
  - "模仿学习"
type:
  - "机器人学"
  - "具身智能"
  - "视觉语言模型"
  - "强化学习"
  - "模仿学习"
  - Robotics
  - "Machine Learning"
arxiv_url: "https://arxiv.org/abs/2503.12533"
pdf_url: "https://arxiv.org/pdf/2503.12533"
code_link: "https://beingbeyond.github.io/being-0"
---

# Being-0: A Humanoid Robotic Agent with Vision-Language Models and Modular Skills

> Building autonomous robotic agents capable of achieving human-level performance in real-world embodied tasks is an ultimate goal in humanoid robot research. Recent advances have made significant progress in high-level cognition with Foundation Models (FMs)…

「Being-0: A Humanoid Robotic Agent with Vision-Language Models and Modular Skills」是 ModelScope 魔搭社区收录的论文，arXiv 2503.12533，作者为 Haoqi Yuan, Yu Bai, Yuhui Fu et al.，发表于 2025-05-11，属于 机器人学、具身智能、视觉语言模型 领域。

- **ArXiv**: 2503.12533
- **Published**: 2025-05-11
- **Authors**: Haoqi Yuan, Yu Bai, Yuhui Fu, Bohan Zhou, Yicheng Feng, Xinrun Xu, Yi Zhan, Börje F. Karlsson, Zongqing Lu
- **Model**: Being-0
- **Developer**: 北京大学、北京智源人工智能研究院、BeingBeyond
- **Domain**: 机器人学, 具身智能, 视觉语言模型, 强化学习, 模仿学习
- **ArXiv URL**: https://arxiv.org/abs/2503.12533
- **PDF**: https://arxiv.org/pdf/2503.12533
- **Code**: https://beingbeyond.github.io/being-0

Source: https://www.modelscope.cn/papers/127565

---

> Being-0：一种结合视觉语言模型与模块化技能的人形机器人智能体

## 摘要

本文提出了 Being-0，一个面向全尺寸人形机器人的分层智能体框架。该框架由三个核心组件构成：用于高层任务规划与推理的基础模型（FM）、作为中间层将语言计划转化为可执行技能指令的轻量级视觉语言模型连接器（Connector），以及提供稳健运动控制和灵巧操作能力的模块化技能库。Being-0 通过云端基础模型与本地计算设备的混合部署策略，有效解决了长视野任务中复合误差和模块延迟问题，在真实世界大型室内环境中实现了复杂的导航与多步操作任务。

## Abstract

Building autonomous robotic agents capable of achieving human-level performance in real-world embodied tasks is an ultimate goal in humanoid robot research. Recent advances have made significant progress in high-level cognition with Foundation Models (FMs) and low-level skill development for humanoid robots. However, directly combining these components often results in poor robustness and efficiency due to compounding errors in long-horizon tasks and the varied latency of different modules. We introduce Being-0, a hierarchical agent framework that integrates an FM with a modular skill library. The FM handles high-level cognitive tasks such as instruction understanding, task planning, and reasoning, while the skill library provides stable locomotion and dexterous manipulation for low-level control. To bridge the gap between these levels, we propose a novel Connector module, powered by a lightweight vision-language model (VLM). The Connector enhances the FM's embodied capabilities by translating language-based plans into actionable skill commands and dynamically coordinating locomotion and manipulation to improve task success. With all components, except the FM, deployable on low-cost onboard computation devices, Being-0 achieves efficient, real-time performance on a full-sized humanoid robot equipped with dexterous hands and active vision. Extensive experiments in large indoor environments demonstrate Being-0's effectiveness in solving complex, long-horizon tasks that require challenging navigation and manipulation subtasks. For further details and videos, visit https://beingbeyond.github.io/Being-0.