@article{sparklemasteringbasicspatialcapabilities,
  title = {Sparkle: Mastering Basic Spatial Capabilities in Vision Language Models Elicits Generalization to Composite Spatial Reasoning},
  author = {Yihong Tang and Ao Qu and Zhaokai Wang and Dingyi Zhuang and Zhaofeng Wu and Wei Ma and Shenhao Wang and Yunhan Zheng and Zhan Zhao and Jinhua Zhao},
  year = {2024},
  eprint = {2410.16162},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2410.16162v2},
}