# Premover: Fast Vision-Language-Action Control by Acting Before Instructions Are Complete

Source: [Premover: Fast Vision-Language-Action Control by Acting Before Instructions Are Complete](<https://www.sota2.com/research/paper/6a0405dadbc836b87d827870>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-05-14T05:00:13.174Z

Authors: Joonha Park, Jiseung Jeong, Taesik Gong

Published: 2026-05-12T14:10:54.000Z

## Abstract

Vision-Language-Action (VLA) policies are typically evaluated as if the user had finished typing or speaking before the robot begins acting. In real deployment, however, users take several seconds to enter a request, leaving the policy idle for a substantial fraction of the interaction. We introduce Premover, a lightweight module that converts this idle window into useful precomputation. Premover keeps the VLA backbone frozen and attaches two small projection heads, one for image patches, one for language tokens, that map an intermediate layer of the backbone into a shared space. The resulting focus map is supervised by simulator-rendered target-object segmentation masks and applied as a per-patch reweighting of the next step's image tokens. A single scalar readiness threshold, trained jointly from streaming prefixes, decides when the policy should begin acting. On the LIBERO benchmark suite, Premover reduces mean wall-clock time from 34.0 to 29.4 seconds, a 13.6% reduction, while matching the full-prompt baseline's success rate (95.1% vs. 95.0%); naive premoving, by contrast, collapses to 66.4%.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Robot Task Completion on LIBERO (Success Rate and Wall-Clock Time)](<https://www.sota2.com/research/sota/robot-task-completion-on-libero-success-rate-and-wall-clock-time.md>): Robot Task Completion; LIBERO; Success Rate (Mean): 95.1.

- [Robot Task Completion on VLA-arena Level-1 (episodes 15-49)](<https://www.sota2.com/research/sota/robot-task-completion-on-vla-arena-level-1-episodes-15-49.md>): Robot Task Completion; VLA-arena Level-1 (episodes 15-49); Success Rate (Extr.): 25.9.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2605.12160>)

[PDF](<https://arxiv.org/pdf/2605.12160.pdf>)