EmbodiedBrain: Expanding Performance Boundaries of Task Planning for Embodied Intelligence
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_ | 1866909865685811200 |
|---|---|
| author | Zou, Ding Wang, Feifan Ge, Mengyu Fan, Siyuan Zhang, Zongbing Chen, Wei Wang, Lingfeng Hu, Zhongyou Yan, Wenrui Gao, Zhengwei Wang, Hao Jin, Weizhao Zhang, Yu Zhao, Hainan Zhang, Mingliang Xi, Xianxian Zhang, Yaru Li, Wenyuan Gao, Zhengguang Zhu, Yurui |
| author_facet | Zou, Ding Wang, Feifan Ge, Mengyu Fan, Siyuan Zhang, Zongbing Chen, Wei Wang, Lingfeng Hu, Zhongyou Yan, Wenrui Gao, Zhengwei Wang, Hao Jin, Weizhao Zhang, Yu Zhao, Hainan Zhang, Mingliang Xi, Xianxian Zhang, Yaru Li, Wenyuan Gao, Zhengguang Zhu, Yurui |
| contents | The realization of Artificial General Intelligence (AGI) necessitates Embodied AI agents capable of robust spatial perception, effective task planning, and adaptive execution in physical environments. However, current large language models (LLMs) and multimodal LLMs (MLLMs) for embodied tasks suffer from key limitations, including a significant gap between model design and agent requirements, an unavoidable trade-off between real-time latency and performance, and the use of unauthentic, offline evaluation metrics. To address these challenges, we propose EmbodiedBrain, a novel vision-language foundation model available in both 7B and 32B parameter sizes. Our framework features an agent-aligned data structure and employs a powerful training methodology that integrates large-scale Supervised Fine-Tuning (SFT) with Step-Augumented Group Relative Policy Optimization (Step-GRPO), which boosts long-horizon task success by integrating preceding steps as Guided Precursors. Furthermore, we incorporate a comprehensive reward system, including a Generative Reward Model (GRM) accelerated at the infrastructure level, to improve training efficiency. For enable thorough validation, we establish a three-part evaluation system encompassing General, Planning, and End-to-End Simulation Benchmarks, highlighted by the proposal and open-sourcing of a novel, challenging simulation environment. Experimental results demonstrate that EmbodiedBrain achieves superior performance across all metrics, establishing a new state-of-the-art for embodied foundation models. Towards paving the way for the next generation of generalist embodied agents, we open-source all of our data, model weight, and evaluating methods, which are available at https://zterobot.github.io/EmbodiedBrain.github.io. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2510_20578 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | EmbodiedBrain: Expanding Performance Boundaries of Task Planning for Embodied Intelligence Zou, Ding Wang, Feifan Ge, Mengyu Fan, Siyuan Zhang, Zongbing Chen, Wei Wang, Lingfeng Hu, Zhongyou Yan, Wenrui Gao, Zhengwei Wang, Hao Jin, Weizhao Zhang, Yu Zhao, Hainan Zhang, Mingliang Xi, Xianxian Zhang, Yaru Li, Wenyuan Gao, Zhengguang Zhu, Yurui Computer Vision and Pattern Recognition Robotics The realization of Artificial General Intelligence (AGI) necessitates Embodied AI agents capable of robust spatial perception, effective task planning, and adaptive execution in physical environments. However, current large language models (LLMs) and multimodal LLMs (MLLMs) for embodied tasks suffer from key limitations, including a significant gap between model design and agent requirements, an unavoidable trade-off between real-time latency and performance, and the use of unauthentic, offline evaluation metrics. To address these challenges, we propose EmbodiedBrain, a novel vision-language foundation model available in both 7B and 32B parameter sizes. Our framework features an agent-aligned data structure and employs a powerful training methodology that integrates large-scale Supervised Fine-Tuning (SFT) with Step-Augumented Group Relative Policy Optimization (Step-GRPO), which boosts long-horizon task success by integrating preceding steps as Guided Precursors. Furthermore, we incorporate a comprehensive reward system, including a Generative Reward Model (GRM) accelerated at the infrastructure level, to improve training efficiency. For enable thorough validation, we establish a three-part evaluation system encompassing General, Planning, and End-to-End Simulation Benchmarks, highlighted by the proposal and open-sourcing of a novel, challenging simulation environment. Experimental results demonstrate that EmbodiedBrain achieves superior performance across all metrics, establishing a new state-of-the-art for embodied foundation models. Towards paving the way for the next generation of generalist embodied agents, we open-source all of our data, model weight, and evaluating methods, which are available at https://zterobot.github.io/EmbodiedBrain.github.io. |
| title | EmbodiedBrain: Expanding Performance Boundaries of Task Planning for Embodied Intelligence |
| topic | Computer Vision and Pattern Recognition Robotics |
| url | https://arxiv.org/abs/2510.20578 |