The Year in AI Papers
The year/Independent research

Paper 2601.12993

Being-H0.5: Scaling Human-Centric Robot Learning for Cross-Embodiment Generalization

Hao Luo, Ye Wang, Wanpeng Zhang, Sipeng Zheng, Ziheng Xi, Chaoyi Xu, Haiweng Xu, Haoqi Yuan, Chi Zhang, Yiqing Wang, Yicheng Feng, Zongqing Lu

Published
Jan 2026
Research lab
Independent
Citations
53
GitHub
1.1K stars

01 In brief

Summary

Being-H0.5 is a foundational Vision-Language-Action (VLA) model for cross-embodiment generalization, introduced by the BeingBeyond Team.

It uses a human-centric learning paradigm, treating human interaction traces as a universal 'mother tongue' for physical interaction.

The model is trained on UniHand-2.0, the largest embodied pre-training recipe to date, with over 35,000 hours of multimodal data (16,000 hours of human video, 14,000 hours of robot data across 30 embodiments, and 5,000 hours of visual-language data), totaling 120B tokens and 400M samples.

A Unified Action Space maps heterogeneous robot controls into semantically aligned slots, enabling low-resource robots to bootstrap from human data.

Architecturally, it uses a Mixture-of-Transformers with a novel Mixture-of-Flow framework, decoupling shared motor primitives from embodiment-specific experts.

For real-world stability, it introduces Manifold-Preserving Gating and Universal Async Chunking.

Being-H0.5 achieves state-of-the-art results on LIBERO (98.9%) and RoboCasa (53.9%) and demonstrates strong cross-embodiment capabilities on five robotic platforms, including emergent zero-shot transfer to unseen task-embodiment pairs.

02 From the paper

Abstract

We introduce Being-H0.5, a foundational Vision-Language-Action (VLA) model designed for robust cross-embodiment generalization across diverse robotic platforms. While existing VLAs often struggle with morphological heterogeneity and data scarcity, we propose a human-centric learning paradigm that treats human interaction traces as a universal "mother tongue" for physical interaction. To support this, we present UniHand-2.0, the largest embodied pre-training recipe to date, comprising over 35,000 hours of multimodal data across 30 distinct robotic embodiments. Our approach introduces a Unified Action Space that maps heterogeneous robot controls into semantically aligned slots, enabling low-resource robots to bootstrap skills from human data and high-resource platforms. Built upon this human-centric foundation, we design a unified sequential modeling and multi-task pre-training paradigm to bridge human demonstrations and robotic execution. Architecturally, Being-H0.5 utilizes a Mixture-of-Transformers design featuring a novel Mixture-of-Flow (MoF) framework to decouple shared motor primitives from specialized embodiment-specific experts. Finally, to make cross-embodiment policies stable in the real world, we introduce Manifold-Preserving Gating for robustness under sensory shift and Universal Async Chunking to universalize chunked control across embodiments with different latency and control profiles. We empirically demonstrate that Being-H0.5 achieves state-of-the-art results on simulated benchmarks, such as LIBERO (98.9%) and RoboCasa (53.9%), while also exhibiting strong cross-embodiment capabilities on five robotic platforms.