StairVLA: Stage-Aware Hierarchical Action Generation for Vision-Language-Action Models

Key points

• Vision-language-action (VLA) models increasingly rely on diffusion- or flow-matching-based action heads to generate continuous robot actions.1
• However, we observe that the conditioning focus naturally shifts across denoising stages: early stages combine language instructions and visual observations to establish a coarse action trajectory, whereas later stages place greater emphasis on current visual observations for action alignment.1
• A high-level VLA performs early denoising to produce a reusable long-horizon partially denoised action trajectory, while a lightweight refiner operates at a higher frequency to refine local action chunks using the latest observations.1
• On LIBERO, our GR00T-style instantiation improves average success from 96.5% to 97.8% while reducing amortized inference latency from 115.0 ms to 44.2 ms per action chunk.1

Sources (1)

• [1]StairVLA: Stage-Aware Hierarchical Action Generation for Vision-Language-Action Models
  arXiv (AI, ML, NLP, CV, robotics, multi-agent) · Oct 6, 04:53 AM

  Based on this insight, we introduce StairVLA, a stage-aware hierarchical action generation framework that uses partially denoised actions as a natural interface between coarse long-horizon action generation and local refinement.

  Vision-language-action (VLA) models increasingly rely on diffusion- or flow-matching-based action heads to generate continuous robot actions.

Extractive summary: sentences quoted from the sources.