OmniVLA-RL: A Vision-Language-Action Model with Spatial Understanding and Online RL

arXiv:2604.17706v1 Announce Type: new Abstract: 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 significantly outperforms state-of-the-art methods, effectively overcoming the fundamental limitations of current VLA models.

Long-Term Memory for VLA-based Agents in Open-World Task Execution
ViVa: A Video-Generative Value Model for Robot Reinforcement Learning
Reflection-Based Task Adaptation for Self-Improving VLA
AVA-VLA: Improving Vision-Language-Action models with Active Visual Attention
DeepThinkVLA: Enhancing Reasoning Capability of Vision-Language-Action Models
HAMLET: Switch your Vision-Language-Action Model into a History-Aware Policy

Source: arXiv cs.RO | 2026-04-21

Related