NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2026
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866915850494148608 |
|---|---|
| author | Zhu, Ziyue Wu, Shangyang Zhao, Shuai Zhao, Zhiqiu Li, Shengjie Wang, Yi Li, Fang Luo, Haoran |
| author_facet | Zhu, Ziyue Wu, Shangyang Zhao, Shuai Zhao, Zhiqiu Li, Shengjie Wang, Yi Li, Fang Luo, Haoran |
| contents | Vision-Language-Action (VLA) models are formulated to ground instructions in visual context and generate action sequences for robotic manipulation. Despite recent progress, VLA models still face challenges in learning related and reusable primitives, reducing reliance on large-scale data and complex architectures, and enabling exploration beyond demonstrations. To address these challenges, we propose a novel Neuro-Symbolic Vision-Language-Action (NS-VLA) framework via online reinforcement learning (RL). It introduces a symbolic encoder to embedding vision and language features and extract structured primitives, utilizes a symbolic solver for data-efficient action sequencing, and leverages online RL to optimize generation via expansive exploration. Experiments on robotic manipulation benchmarks demonstrate that NS-VLA outperforms previous methods in both one-shot training and data-perturbed settings, while simultaneously exhibiting superior zero-shot generalizability, high data efficiency and expanded exploration space. Our code is available. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2603_09542 |
| institution | arXiv |
| publishDate | 2026 |
| record_format | arxiv |
| spellingShingle | NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models Zhu, Ziyue Wu, Shangyang Zhao, Shuai Zhao, Zhiqiu Li, Shengjie Wang, Yi Li, Fang Luo, Haoran Robotics Vision-Language-Action (VLA) models are formulated to ground instructions in visual context and generate action sequences for robotic manipulation. Despite recent progress, VLA models still face challenges in learning related and reusable primitives, reducing reliance on large-scale data and complex architectures, and enabling exploration beyond demonstrations. To address these challenges, we propose a novel Neuro-Symbolic Vision-Language-Action (NS-VLA) framework via online reinforcement learning (RL). It introduces a symbolic encoder to embedding vision and language features and extract structured primitives, utilizes a symbolic solver for data-efficient action sequencing, and leverages online RL to optimize generation via expansive exploration. Experiments on robotic manipulation benchmarks demonstrate that NS-VLA outperforms previous methods in both one-shot training and data-perturbed settings, while simultaneously exhibiting superior zero-shot generalizability, high data efficiency and expanded exploration space. Our code is available. |
| title | NS-VLA: Towards Neuro-Symbolic Vision-Language-Action Models |
| topic | Robotics |
| url | https://arxiv.org/abs/2603.09542 |