English

Large Vocabulary Size Improves Large Language Models

Computation and Language 2025-05-29 v2

Abstract

This paper empirically investigates the relationship between subword vocabulary size and the performance of large language models (LLMs) to provide insights on how to define the vocabulary size. Experimental results show that larger vocabulary sizes lead to better performance in LLMs. Moreover, we consider a continual training scenario where a pre-trained language model is trained on a different target language. We introduce a simple method to use a new vocabulary instead of the pre-defined one. We show that using the new vocabulary outperforms the model with the vocabulary used in pre-training.

Keywords

Cite

@article{arxiv.2406.16508,
  title  = {Large Vocabulary Size Improves Large Language Models},
  author = {Sho Takase and Ryokan Ri and Shun Kiyono and Takuya Kato},
  journal= {arXiv preprint arXiv:2406.16508},
  year   = {2025}
}

Comments

Findings of ACL 2025

R2 v1 2026-06-28T17:17:05.361Z