AI RESEARCH
Closed-Loop Action Chunks with Dynamic Corrections for Training-Free Diffusion Policy
arXiv CS.AI
•
ArXi:2603.01953v2 Announce Type: replace-cross Diffusion-based policies have achieved remarkable results in robotic manipulation but often struggle to adapt rapidly in dynamic scenarios, leading to delayed responses or task failures. We present DCDP, a Dynamic Closed-Loop Diffusion Policy framework that integrates chunk-based action generation with real-time correction.