Implicit Drifting Policy: One-Step Action Generation via Conditional Expert Geometry

ArXi:2606.01098v1 Announce Type: cross Generative action policies based on diffusion or flow matching excel in behavior cloning, yet their iterative sampling is prohibitive for high-frequency robot control. While recent one-step formulations alleviate this latency, they inevitably discard the intermediate trajectory evolution that provides crucial action correction. Directly recovering this mechanism by explicitly estimating a