Realtime-VLA FLASH: Speculative Inference Framework for Diffusion-based VLAs
摘要
Diffusion-based vision-language-action models (dVLAs) are promising for embodied intelligence but are fundamentally limited in real-time deployment by the high latency of full inference. We propose Realtime-VLA FLASH, a speculative inference framework that eliminates most full inference calls during replanning by introducing a lightweight draft model with parallel verification via the main model's Action Expert and a phase-aware fallback mechanism that reverts to the full inference pipeline when needed. This design enables low-latency, high-frequency replanning without sacrificing reliability. Experiments show that on LIBERO, FLASH largely preserves task performance by replacing many 58.0 ms full-inference rounds with speculative rounds as fast as 7.8 ms, lowering task-level average inference latency to 19.1 ms (3.04x speedup). We additionally demonstrate effectiveness on real-world conveyor-belt sorting, highlighting its practical impact for latency-critical embodied tasks.
引用
@article{arxiv.2605.13778,
title = {Realtime-VLA FLASH: Speculative Inference Framework for Diffusion-based VLAs},
author = {Jiahui Niu and Kefan Gu and Yucheng Zhao and Shengwen Liang and Tiancai Wang and Xing Hu and Ying Wang and Huawei Li},
journal= {arXiv preprint arXiv:2605.13778},
year = {2026}
}