GROVE: A Generalized Reward for Learning Open-Vocabulary Physical Skill
Fuente:
arXiv
Guardado en:
| Autores principales: | , , , , , , , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2025
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866915490789588992 |
|---|---|
| author | Cui, Jieming Liu, Tengyu Meng, Ziyu Yu, Jiale Song, Ran Zhang, Wei Zhu, Yixin Huang, Siyuan |
| author_facet | Cui, Jieming Liu, Tengyu Meng, Ziyu Yu, Jiale Song, Ran Zhang, Wei Zhu, Yixin Huang, Siyuan |
| contents | Learning open-vocabulary physical skills for simulated agents presents a significant challenge in artificial intelligence. Current reinforcement learning approaches face critical limitations: manually designed rewards lack scalability across diverse tasks, while demonstration-based methods struggle to generalize beyond their training distribution. We introduce GROVE, a generalized reward framework that enables open-vocabulary physical skill learning without manual engineering or task-specific demonstrations. Our key insight is that Large Language Models(LLMs) and Vision Language Models(VLMs) provide complementary guidance -- LLMs generate precise physical constraints capturing task requirements, while VLMs evaluate motion semantics and naturalness. Through an iterative design process, VLM-based feedback continuously refines LLM-generated constraints, creating a self-improving reward system. To bridge the domain gap between simulation and natural images, we develop Pose2CLIP, a lightweight mapper that efficiently projects agent poses directly into semantic feature space without computationally expensive rendering. Extensive experiments across diverse embodiments and learning paradigms demonstrate GROVE's effectiveness, achieving 22.2% higher motion naturalness and 25.7% better task completion scores while training 8.4x faster than previous methods. These results establish a new foundation for scalable physical skill acquisition in simulated environments. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2504_04191 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | GROVE: A Generalized Reward for Learning Open-Vocabulary Physical Skill Cui, Jieming Liu, Tengyu Meng, Ziyu Yu, Jiale Song, Ran Zhang, Wei Zhu, Yixin Huang, Siyuan Computer Vision and Pattern Recognition Robotics Learning open-vocabulary physical skills for simulated agents presents a significant challenge in artificial intelligence. Current reinforcement learning approaches face critical limitations: manually designed rewards lack scalability across diverse tasks, while demonstration-based methods struggle to generalize beyond their training distribution. We introduce GROVE, a generalized reward framework that enables open-vocabulary physical skill learning without manual engineering or task-specific demonstrations. Our key insight is that Large Language Models(LLMs) and Vision Language Models(VLMs) provide complementary guidance -- LLMs generate precise physical constraints capturing task requirements, while VLMs evaluate motion semantics and naturalness. Through an iterative design process, VLM-based feedback continuously refines LLM-generated constraints, creating a self-improving reward system. To bridge the domain gap between simulation and natural images, we develop Pose2CLIP, a lightweight mapper that efficiently projects agent poses directly into semantic feature space without computationally expensive rendering. Extensive experiments across diverse embodiments and learning paradigms demonstrate GROVE's effectiveness, achieving 22.2% higher motion naturalness and 25.7% better task completion scores while training 8.4x faster than previous methods. These results establish a new foundation for scalable physical skill acquisition in simulated environments. |
| title | GROVE: A Generalized Reward for Learning Open-Vocabulary Physical Skill |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2504.04191 |