Scaling Behavior Foundation Model for Humanoid Robots

Weishuai Zeng, Kangning Yin, Xiaojie Niu, Shunlin Lu, Weixiang Zhong, Jiahe Chen, Feiyu Jia, Xiao Chen, Zirui Wang, Furui Xu, Ming Zhou, Kailin Li, Weinan Zhang, He Wang, Li Yi, Dahua Lin, Jiangmiao Pang, Jingbo Wang · N/A · 2026

Summary

Abstract Summary

Humanoid control requires natural whole-body coordination, precise real-time responses to control signals, and robust generalization across diverse environmental contexts, making it a cornerstone for generalist embodied agents. Behavior Foundation Models (BFMs) have recently emerged as a promising solution to address these challenges by leveraging large-scale behavioral data to achieve superior expressiveness, versatility and generalization. However, despite growing interest in scaling BFMs to further improve their capabilities, it remains unclear how key factors, including the learning paradigm, behavioral data and model architecture should be coordinated to enable effective scaling. In this work, we revisit the scaling recipe for BFMs and demonstrate that substantial performance gains can be achieved through the coordination of three core components: 1) the learning paradigm of motion tracking that reformulates diverse humanoid control problems as the reproduction of integrated whole-body behaviors in the global frame; 2) the strategic synergy between on-policy rollout quantity and reference motion diversity; and 3) the expressive and scalable model architecture termed Humanoid Transformer that facilitates the natural emergence of structured behavioral representations. Through extensive experiments in both simulation and real-world deployment, we demonstrate that our approach yields significant improvements in control fidelity and task generalization, reducing Mean Per-Keypoint Position Error (MPKPE) on the test set by over 10% in local mode and 82% in global mode compared with existing humanoid controllers. These results establish BFM as a principled and effective foundation for scalable and general-purpose humanoid control.

Key Points

Humanoid control requires natural whole-body coordination, precise real-time responses to control...
Behavior Foundation Models (BFMs) have recently emerged as a promising solution to address these ...
However, despite growing interest in scaling BFMs to further improve their capabilities, it remai...
In this work, we revisit the scaling recipe for BFMs and demonstrate that substantial performance...
Through extensive experiments in both simulation and real-world deployment, we demonstrate that o...

Scaling Behavior Foundation Model for Humanoid Robots

|Authors: Weishuai Zeng, Kangning Yin, Xiaojie Niu, Shunlin Lu, Weixiang Zhong, Jiahe Chen, Feiyu Jia, Xiao Chen, Zirui Wang, Furui Xu, Ming Zhou, Kailin Li, Weinan Zhang, He Wang, Li Yi, Dahua Lin, Jiangmiao Pang, Jingbo Wang

|Venue: arXiv preprint | Year: 2026

|arXiv: 2607.15163v1

Abstract

Key Contributions

Humanoid control requires natural whole-body coordination, precise real-time responses to control…
Behavior Foundation Models (BFMs) have recently emerged as a promising solution to address these …
However, despite growing interest in scaling BFMs to further improve their capabilities, it remai…
In this work, we revisit the scaling recipe for BFMs and demonstrate that substantial performance…
Through extensive experiments in both simulation and real-world deployment, we demonstrate that o…

Topics

reinforcement-learning
control

Code & Data

GitHub repository: https://github.com/NVLabs/ProtoMotions

BibTeX

@article{Zeng2026_260715163v1,
  title     = {Scaling Behavior Foundation Model for Humanoid Robots},
  author    = {Weishuai Zeng and Kangning Yin and Xiaojie Niu and Shunlin Lu and Weixiang Zhong and Jiahe Chen and Feiyu Jia and Xiao Chen and Zirui Wang and Furui Xu and Ming Zhou and Kailin Li and Weinan Zhang and He Wang and Li Yi and Dahua Lin and Jiangmiao Pang and Jingbo Wang},
  year      = {2026},
  eprint    = {2607.15163v1},
  archivePrefix = {arXiv},
  primaryClass  = {cs.RO},
  url       = {https://arxiv.org/abs/2607.15163v1}
}