# OGPO: Offline Goal-conditioned Policy Optimization for Recoverable Vision-Language-Action Models

**Published:** October 12, 2026
**Author:** Xule Gao, Xi Wang, Zehua Zang, Rui Wang, Changwen Zheng, Chuxiong Sun
**Source:** arXiv — cs.AI
**Language:** en-US
**HTML:** https://aiqazaqstan.com/en/articles/ogpo-offline-goal-conditioned-policy-optimization-for-recoverable-vision

arXiv:2610.12914v1 Announce Type: new Abstract: Vision-language-action (VLA) models have shown strong promise in robotic manipulation, yet their post-training still predominantly relies on trajectory-level supervised imitation over robot demonstrations. While effective for…

> From the original source: **arXiv — cs.AI**
> [Read full article at arXiv — cs.AI](https://arxiv.org/abs/2610.12914)

**Tags:** #Робототехника #arXiv #OpenAI Robotics

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*