Humanoid World Models: Open World Foundation Models for Humanoid Robotics
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909680696033280 |
|---|---|
| author | Ali, Muhammad Qasim Sridhar, Aditya Matiana, Shahbuland Wong, Alex Al-Sharman, Mohammad |
| author_facet | Ali, Muhammad Qasim Sridhar, Aditya Matiana, Shahbuland Wong, Alex Al-Sharman, Mohammad |
| contents | Humanoid robots, with their human-like form, are uniquely suited for interacting in environments built for people. However, enabling humanoids to reason, plan, and act in complex open-world settings remains a challenge. World models, models that predict the future outcome of a given action, can support these capabilities by serving as a dynamics model in long-horizon planning and generating synthetic data for policy learning. We introduce Humanoid World Models (HWM), a family of lightweight, open-source models that forecast future egocentric video conditioned on humanoid control tokens. We train two types of generative models, Masked Transformers and Flow-Matching, on 100 hours of humanoid demonstrations. Additionally, we explore architectural variants with different attention mechanisms and parameter-sharing strategies. Our parameter-sharing techniques reduce model size by 33-53% with minimal impact on performance or visual fidelity. HWMs are designed to be trained and deployed in practical academic and small-lab settings, such as 1-2 GPUs. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2506_01182 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Humanoid World Models: Open World Foundation Models for Humanoid Robotics Ali, Muhammad Qasim Sridhar, Aditya Matiana, Shahbuland Wong, Alex Al-Sharman, Mohammad Robotics Artificial Intelligence Humanoid robots, with their human-like form, are uniquely suited for interacting in environments built for people. However, enabling humanoids to reason, plan, and act in complex open-world settings remains a challenge. World models, models that predict the future outcome of a given action, can support these capabilities by serving as a dynamics model in long-horizon planning and generating synthetic data for policy learning. We introduce Humanoid World Models (HWM), a family of lightweight, open-source models that forecast future egocentric video conditioned on humanoid control tokens. We train two types of generative models, Masked Transformers and Flow-Matching, on 100 hours of humanoid demonstrations. Additionally, we explore architectural variants with different attention mechanisms and parameter-sharing strategies. Our parameter-sharing techniques reduce model size by 33-53% with minimal impact on performance or visual fidelity. HWMs are designed to be trained and deployed in practical academic and small-lab settings, such as 1-2 GPUs. |
| title | Humanoid World Models: Open World Foundation Models for Humanoid Robotics |
| topic | Robotics Artificial Intelligence |
| url | https://arxiv.org/abs/2506.01182 |