Retrieve to Act: Motion Primitive Graph Retrieval for Robust VLA Execution
Retrieve to Act: Motion Primitive Graph Retrieval for Robust VLA Execution
This research proposes R2A, a framework that enhances the robustness of Vision-Language-Action (VLA) policies by treating out-of-distribution visual shifts as a geometric motion primitive retrieval problem. R2A uses a graph neural network to dynamically localize robot execution and harmonize retrieved geometric priors with predicted actions, achieving significant improvements in simulation and real-world benchmarks without policy finetuning.
This paper introduces R2A, a test-time scaling framework for improving the robustness of Vision-Language-Action (VLA) policies against out-of-distribution visual shifts. By leveraging motion primitive graphs and a learned gating mechanism, R2A enhances performance in both simulation and real-world settings without requiring policy finetuning.
Source: arXiv Robotics — research abstracts · Read original article ↗
Loading article text…
Source：arXiv Robotics — research abstracts · arxiv.org