SDN: Improving VLA policy robustness at test time via selected diffusion noise

The authors propose Selected Diffusion Noise (SDN), a training-free method that treats diffusion noise as a controllable degree of freedom to improve the test-time reliability of VLA policies. SDN samples multiple candidate action chunks and performs two-stage selection: it first filters by task-relevant visual grounding using a counterfactual negative set, then selects the kinematically most stable action chunk based on JerkRMS. In simulation and real-world experiments with three VLA policies, it achieved improvements of 8-10% and 18.33%, respectively.