AI RESEARCH
DFM-VLA: Iterative Action Refinement for Robot Manipulation via Discrete Flow Matching
arXiv CS.CV
•
ArXi:2603.26320v1 Announce Type: cross Vision--Language--Action (VLA) models that encode actions using a discrete tokenization scheme are increasingly adopted for robotic manipulation, but existing decoding paradigms remain fundamentally limited. Whether actions are decoded sequentially by autoregressive VLAs or in parallel by discrete diffusion VLAs, once a token is generated, it is typically fixed and cannot be revised in subsequent iterations, so early token errors cannot be effectively corrected later.