@inproceedings{limitstodepthefficienciesofself,
  title = {The Depth-to-Width Interplay in Self-Attention},
  author = {Yoav Levine and Noam Wies and Or Sharir and Hofit Bata and Amnon Shashua},
  year = {2020},
  booktitle = {NeurIPS 2020 12},
  eprint = {2006.12467},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2006.12467v3},
}