← Back to Benchmarks
simmediumroboticsmetric · varies

ProbeFlow: Training-Free Adaptive Flow Matching for Vision-Language-Action Models

Description

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 fr

Source

http://arxiv.org/abs/2603.17850v1