@misc{indiciaebe90f43d1c73,
  title = {Rethinking and Improving Relative Position Encoding for Vision Transformer},
  author = {Kan Wu and Houwen Peng and Minghao Chen and Jianlong Fu and Hongyang Chao},
  year = {2021},
  url = {https://arxiv.org/abs/2107.14222},
  note = {Source identifier: 2107.14222}
}