@article{inconsistenttokenizationscauselanguage, title = {Inconsistent Tokenizations Cause Language Models to be Perplexed by Japanese Grammar}, author = {Andrew Gambardella and Takeshi Kojima and Yusuke Iwasawa and Yutaka Matsuo}, year = {2025}, eprint = {2505.19599}, archivePrefix = {arXiv}, url = {https://arxiv.org/abs/2505.19599v1}, }