@article{codebpeinvestigatingsubtokenizationoptio, title = {CodeBPE: Investigating Subtokenization Options for Large Language Model Pretraining on Source Code}, author = {Nadezhda Chirkova and Sergey Troshin}, year = {2023}, eprint = {2308.00683}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2308.00683v1}, }