Saved in:
| Main Authors: | , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2509.12437 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908540690497536 |
|---|---|
| author | Wang, Dingrui Sun, Zhexiao Li, Zhouheng Wang, Cheng Peng, Youlun Ye, Hongyuan Zarrouki, Baha Li, Wei Piccinini, Mattia Xie, Lei Betz, Johannes |
| author_facet | Wang, Dingrui Sun, Zhexiao Li, Zhouheng Wang, Cheng Peng, Youlun Ye, Hongyuan Zarrouki, Baha Li, Wei Piccinini, Mattia Xie, Lei Betz, Johannes |
| contents | A major challenge in deploying world models is the trade-off between size and performance. Large world models can capture rich physical dynamics but require massive computing resources, making them impractical for edge devices. Small world models are easier to deploy but often struggle to learn accurate physics, leading to poor predictions. We propose the Physics-Informed BEV World Model (PIWM), a compact model designed to efficiently capture physical interactions in bird's-eye-view (BEV) representations. PIWM uses Soft Mask during training to improve dynamic object modeling and future prediction. We also introduce a simple yet effective technique, Warm Start, for inference to enhance prediction quality with a zero-shot model. Experiments show that at the same parameter scale (400M), PIWM surpasses the baseline by 60.6% in weighted overall score. Moreover, even when compared with the largest baseline model (400M), the smallest PIWM (130M Soft Mask) achieves a 7.4% higher weighted overall score with a 28% faster inference speed. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2509_12437 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Enhancing Physical Consistency in Lightweight World Models Wang, Dingrui Sun, Zhexiao Li, Zhouheng Wang, Cheng Peng, Youlun Ye, Hongyuan Zarrouki, Baha Li, Wei Piccinini, Mattia Xie, Lei Betz, Johannes Artificial Intelligence A major challenge in deploying world models is the trade-off between size and performance. Large world models can capture rich physical dynamics but require massive computing resources, making them impractical for edge devices. Small world models are easier to deploy but often struggle to learn accurate physics, leading to poor predictions. We propose the Physics-Informed BEV World Model (PIWM), a compact model designed to efficiently capture physical interactions in bird's-eye-view (BEV) representations. PIWM uses Soft Mask during training to improve dynamic object modeling and future prediction. We also introduce a simple yet effective technique, Warm Start, for inference to enhance prediction quality with a zero-shot model. Experiments show that at the same parameter scale (400M), PIWM surpasses the baseline by 60.6% in weighted overall score. Moreover, even when compared with the largest baseline model (400M), the smallest PIWM (130M Soft Mask) achieves a 7.4% higher weighted overall score with a 28% faster inference speed. |
| title | Enhancing Physical Consistency in Lightweight World Models |
| topic | Artificial Intelligence |
| url | https://arxiv.org/abs/2509.12437 |