# Fast-VLA preprint reports faster robot control by reusing task memory

_Published Thursday, October 15, 2026 at 9:34 PM EDT · Science, Robotics · Latest · Tier 2 — Notable_

Researchers report that Fast-VLA reduces robot policy inference time by 10.9-17.1x against three continually conditioned baselines on the same hardware in RoboTwin evaluations. The preprint describes a robot control policy that runs a frozen 4B vision-language model once per episode to store instructions and initial views as task memory.

A lightweight actor then combines current visual and robot-state feedback with that memory to predict actions without rerunning the foundation model. Fast-VLA achieves 92.46% and 92.40% success on RoboTwin 2.0 Clean and Randomized, respectively, and requires 1.22 GiB of online resident memory after initial processing. The reported speed comparison is limited to RoboTwin evaluations.

## Sources

- [arXiv Query: search_query=cat:cs.RO&id_list=&start=0&max_results=30](https://arxiv.org/abs/2610.21389v1)

---
Canonical: https://techandbusiness.org/newswire/9QkrOGkZAVc3UtLNbE9GUh
Published: 2026-10-16T01:34:30.049Z
Story chronology: 2026-10-15T07:05:16.000Z
Retrieved: 2026-10-16T04:51:58.950Z
Publisher: Tech & Business (techandbusiness.org)