@article{stemvlaanopensourcevisionlanguageactionm,
  title = {StemVLA:An Open-Source Vision-Language-Action Model with Future 3D Spatial Geometry Knowledge and 4D Historical Representation},
  author = {Jiasong Xiao and Yutao She and Kai Li and Yuyang Sha and Ziang Cheng},
  year = {2026},
  eprint = {2602.23721},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2602.23721},
}