# Let It Be Simple: One-Step Action Generation for Vision-Language-Action Models

Source: [Let It Be Simple: One-Step Action Generation for Vision-Language-Action Models](<https://www.sota2.com/research/paper/6a22589c86173ad9c1468b8d>)

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

Data updated: 2026-07-15T05:00:05.882Z

Authors: Yitong Chen, Shiduo Zhang, Jingjing Gong, Xipeng Qiu

Published: 2026-06-04T05:58:30.000Z

## Abstract

Generating diverse images from sparse text is hard; generating compact actions from rich observations is easier. From the condition-target view, Vision-Language-Action (VLA) thus aligns with image-to-text, not text-to-image. We formalize this view through the irreducible velocity loss $R\_v(t,c)$ of standard flow matching and validate it with a controlled 8-mode toy experiment and image-to-text MNIST task. We then show that high-noise training boosts one-step VLA decoding on standard LIBERO, achieving 95.6% on LIBERO-Long, and remains competitive across LIBERO-Plus, LIBERO-Pro, and real-world robot tasks, while ablations that weaken the condition or expand the horizon predictably erase the one-step gain. These results suggest that whether one-step action generation works in VLA depends not on specialized training, but on the condition-target structure.

## 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 Manipulation on Bimanual YAM RSS OpenPI pi0.5 Policies](<https://www.sota2.com/research/sota/robot-manipulation-on-bimanual-yam-rss-openpi-pi0-5-policies.md>): Robot Manipulation; Bimanual YAM RSS Insert mouse battery real-robot specialist policies OpenPI pi0.5; Success Rate: 80.

- [Robot Manipulation on Bimanual YAM RSS Seal water bottle cap OpenPI pi0.5 (real-robot specialist policies)](<https://www.sota2.com/research/sota/robot-manipulation-on-bimanual-yam-rss-seal-water-bottle-cap-openpi-pi0-5-real-robot-specialist-policies.md>): Robot Manipulation; Bimanual YAM RSS Seal water bottle cap OpenPI pi0.5 (real-robot specialist policies); Success Rate: 60.

- [Robot Manipulation on Bimanual YAM RSS Tower of Hanoi game OpenPI pi0.5](<https://www.sota2.com/research/sota/robot-manipulation-on-bimanual-yam-rss-tower-of-hanoi-game-openpi-pi0-5.md>): Robot Manipulation; Bimanual YAM RSS Tower of Hanoi game real-robot specialist policies OpenPI pi0.5; Success Rate: 100.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2606.05737>)

[PDF](<https://arxiv.org/pdf/2606.05737.pdf>)