Visual-Language-Action (VLA) models represent a paradigm shift in embodied AI, yet existing frameworks often struggle with imprecise spatial perception, suboptimal multimodal fusion, and instability in reinforcement learning. To bridge these gaps, we propose OmniVLA-RL, a novel architecture that leverages a Mix-of-Transformers (MoT) design to synergistically integrate reasoning, spatial, and action experts. Furthermore, we introduce Flow-GSPO, which reformulates flow matching as a Stochastic Differential Equation (SDE) process and integrates it with Group Segmented Policy Optimization (GSPO) to enhance action precision and training robustness. Extensive evaluations on the LIBERO and LIBERO-Plus benchmarks demonstrate that OmniVLA-RL achieves decent overall performance and surpasses mainstream existing methods, effectively overcoming the fundamental limitations of current VLA models.
@article{arxiv.2604.17706,
title = {OmniVLA-RL: A Vision-Language-Action Model with Spatial Understanding and Online RL},
author = {Haoxiang Jie and Yaoyuan Yan and Xiangyu Wei and Kailin Wang and Hongjie Yan and Zhiyou Heng and Daocheng Chen},
journal= {arXiv preprint arXiv:2604.17706},
year = {2026}
}