ProbeFlow: Training-Free Adaptive Flow Matching for Vision-Language-Action Models
This addresses the action head bottleneck for low-latency continuous robotic control, offering a practical solution for responsive physical deployments, though it is incremental as it optimizes an existing method rather than introducing a new paradigm.
The paper tackled the inference latency problem in Vision-Language-Action models with Flow Matching action heads by proposing ProbeFlow, a training-free adaptive inference framework that dynamically schedules integration steps, achieving a 14.8x acceleration in action decoding and a 2.8x reduction in end-to-end system latency on the MetaWorld benchmark without compromising success rates.
Recent Vision-Language-Action (VLA) models equipped with Flow Matching (FM) action heads achieve state-of-the-art performance in complex robot manipulation. However, the multi-step iterative ODE solving required by FM introduces inference latency that precludes responsive physical control. While current acceleration efforts optimize the Vision-Language Model (VLM) backbone, the action head bottleneck remains overlooked. To address this, we propose ProbeFlow, a training-free adaptive inference framework tai- lored for continuous robotic control. By evaluating geometric trajectory complexity via the cosine similarity between initial and lookahead velocity vectors, ProbeFlow dynamically sched- ules integration steps to prune redundant network evaluations. On the MetaWorld benchmark, it accelerates action decoding by 14.8x (reducing average steps from N = 50 to 2.6) and cuts end-to-end system latency by 2.8x without compromising the manipulation success rate. On the long-horizon LIBERO benchmark, the probe automatically allocates a denser schedule to navigate semantic bottlenecks, effectively resolving the flow solver delay. Real-world physical deployments confirm that ProbeFlow successfully mitigates action decoding latency while ensuring execution stability, offering a highly practical solution for low-latency continuous generative policies.