List your company

Cross-language Vocabulary Overlap on 6 languages Different script

0.77JSDUnigram
0.76080.82290.8850.9471May 26, 2023

Evaluation results

MethodLinks
UnigramTokenizer=Unigram2023.050.77
TokMixTokenizer=TokMix2023.050.8
BPETokenizer=BPE2023.050.83
NoOverlapTokenizer=NoOverlap2023.051