Kinematic MeanFlow: One-Step Action Generation Policy for Robotic Foundation Models

Researchers studying one-step action generation in Robotic Foundation Models found that directly applying the MeanFlow framework causes performance collapse, according to a paper titled "Kinematic MeanFlow: One-Step Action Generation Policy for Robotic Foundation Models." The work targets the high inference latency of multi-step flow matching in robotic foundation models.

In this paper, we study how to achieve one-step action generation in Robotic Foundation Models RFMs , aiming to overcome the high inference latency of multi-step flow matching. MeanFlow provides a promising framework for this goal, yet its direct application leads to performance collapse. We discov