Paper ID: 2409.04599

BPE Gets Picky: Efficient Vocabulary Refinement During Tokenizer Training

Pavel Chizhov, Catherine Arnett, Elizaveta Korotkova, Ivan P. Yamshchikov

Language models can largely benefit from efficient tokenization. However, they still mostly utilize the classical BPE algorithm, a simple and reliable method. This has been shown to cause such issues as under-trained tokens and sub-optimal compression that may affect the downstream performance. We introduce Picky BPE, a modified BPE algorithm that carries out vocabulary refinement during tokenizer training. Our method improves vocabulary efficiency, eliminates under-trained tokens, and does not compromise text compression. Our experiments show that our method does not reduce the downstream performance, and in several cases improves it.

Submitted: Sep 6, 2024