SOURCE-LINKED INTELLIGENCE

Fewer Steps, Better Actions: Rethinking Flow-Matching Inference for VLA Policies

arXiv · AI, language, vision and robotics · article · Sep 18, 2026 · UTC

Vision-language-action (VLA) policies based on flow matching generate action chunks through repeated evaluations of an action expert. Increasing the number of integration steps raises inference cost, but does not necessarily improve closed-loop success. We propose Coda, which reallocates part of this integration budget to a single learned endpoint correction. A frozen policy first completes a few-step noise-to-action trajectory; a lightweight Transformer then predicts a demonstration-supervised residual using the candidate action, source noise, and shared observation-prefix cache. Only the cor

Read original source ↗ Open in workspace

recordType

paper

region

Global