@inproceedings{bytepairencodingissuboptimalforlanguage,
  title = {Byte Pair Encoding is Suboptimal for Language Model Pretraining},
  author = {Kaj Bostrom and Greg Durrett},
  year = {2020},
  booktitle = {Findings of the Association for Computational Linguistics 2020},
  eprint = {2004.03720},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2004.03720v2},
}