Saved in:
Bibliographic Details
Main Authors: Wolleb, Benoist, Silvestri, Romain, Vernikos, Giorgos, Dolamic, Ljiljana, Popescu-Belis, Andrei
Format: Preprint
Published: 2023
Subjects:
Online Access:https://arxiv.org/abs/2306.01393
Tags: Add Tag
No Tags, Be the first to tag this record!
Table of Contents:
  • Subword tokenization is the de facto standard for tokenization in neural language models and machine translation systems. Three advantages are frequently cited in favor of subwords: shorter encoding of frequent tokens, compositionality of subwords, and ability to deal with unknown words. As their relative importance is not entirely clear yet, we propose a tokenization approach that enables us to separate frequency (the first advantage) from compositionality. The approach uses Huffman coding to tokenize words, by order of frequency, using a fixed amount of symbols. Experiments with CS-DE, EN-FR and EN-DE NMT show that frequency alone accounts for 90%-95% of the scores reached by BPE, hence compositionality has less importance than previously thought.