# GPL: Representing VLA reasoning state as a plan distribution

- Source: arXiv cs.RO
- Published: 2026-10-21T04:00:00.000Z
- RoboHot: https://robohot.news/en/items/yruwdhw16arr725tkti0ard5v
- Original: https://arxiv.org/abs/2610.28166

## Summary

The authors propose Generative Plan Latents (GPL), which represent the reasoning state of vision-language-action (VLA) models as a low-dimensional conditional distribution over plans, such as which object, grasp, or route to use, so that plans can be sampled, selected on request, and reweighted by deployment outcomes without retraining the VLA.