STEP: Warm-Started Visuomotor Policies with Spatiotemporal Consistency Prediction

ArXi:2602.08245v2 Announce Type: replace-cross Diffusion policies have recently emerged as a powerful paradigm for visuomotor control in robotic manipulation due to their ability to model the distribution of action sequences and capture multimodality. However, iterative denoising leads to substantial inference latency, limiting control frequency in real-time closed-loop systems.