Multilingual Word Segmentation: Training Many Language-Specific Tokenizers Smoothly Thanks to the Universal Dependencies Corpus.

Erwan Moreau, Carl Vogel

Browse the full LREC paper archive.