Article

VLA-Rolling: Fewer Denoising Steps and Better Control via Adaptive Action Reuse

VLA-Rolling is a lightweight inference framework that reuses action tails through three stages: action-tail inheritance, per-action confidence estimation, and confidence-guided refinement based on renoising and denoising step allocation, while the underlying VLA model remains frozen.

Part of the event: VLA-Rolling reuses action tails to reduce denoising steps, achieving up to 1.48× speedup across 16 experiment settings

This page provides a summary and source link only.