@article{busefficientandeffectivevisionlanguage,
  title = {BUS:Efficient and Effective Vision-language Pre-training with Bottom-Up Patch Summarization},
  author = {Chaoya Jiang and Haiyang Xu and Wei Ye and Qinghao Ye and Chenliang Li and Ming Yan and Bin Bi and Shikun Zhang and Fei Huang and Songfang Huang},
  year = {2023},
  eprint = {2307.08504},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2307.08504v2},
}