Towards Deploying VLA without Fine-Tuning: Plug-and-Play Inference-Time VLA Policy Steering via Embodied Evolutionary Diffusion

Vision-Language-Action (VLA) models have demonstrated significant potential in real-world robotic manipulation. However, pre-trained VLA policies still suffer from substantial performance degradation during downstream deployment. Although fine-tuning can mitigate this issue, its reliance on costly demonstration collection and intensive computation makes it impractical in real-world settings. In this work, we introduce VLA-Pilot, a plug-and-play inference-time policy steering method for zero-shot deployment of pre-trained VLA without any additional fine-tuning or data collection. We evaluate VLA-Pilot on both simulation and real-world experiments across distinct robotic embodiments. Experimental results demonstrate that VLA-Pilot substantially boosts the success rates of off-the-shelf pre-trained VLA policies, enabling robust zero-shot generalization to diverse downstream tasks and embodiments.