Think Twice, Act Once: Token-Aware Compression and Action Reuse for Efficient Inference in Vision-Language-Action Models

Understand

Vision-Language-Action (VLA) models have emerged as a powerful paradigm for general-purpose robot control through natural language instructions.

• However, their high inference cost-stemming from large-scale token computation and autoregressive decoding-poses significant challenges for real-time deployment and edge applications.
• While prior work has primarily focused on architectural optimization, we take a different perspective by identifying a dual form of redundancy in VLA models: (i) high similarity across consecutive action steps, and (ii) substantial redundancy in visual tokens.
• Motivated by these observations, we propose FlashVLA, the first training-free and plug-and-play acceleration framework that enables action reuse in VLA models.