Scaling Sim-to-Real VLA Reinforcement Learning with Generative 3D Worlds
Uses 3D world generative models to build 100 diverse interactive scenes for RL fine-tuning of a VLA. Reports sim success from 9.7% to 79.8% and real-world success from 21.7% to 75%.用 3D 世界生成模型构建 100 个多样的可交互场景，对 VLA 做强化学习微调。报告仿真成功率从 9.7% 提升到 79.8%，真机从 21.7% 提升到 75%。
BibTeX
@misc{choi2026scaling,
title = {{Scaling Sim-to-Real VLA Reinforcement Learning with Generative 3D Worlds}},
author = {Andrew Choi and Xinjie Wang and Zhizhong Su and Wei Xu},
year = {2026},
eprint = {2603.18532},
archivePrefix = {arXiv},
primaryClass = {cs.RO},
url = {https://arxiv.org/abs/2603.18532},
}