BPEmb: Tokenization-free Pre-trained Subword Embeddings in 275 Languages.

Benjamin Heinzerling Michael Strube

Published in: LREC (2018)

Keyphrases

pre trained
n gram
character n grams
training data
control signals
manifold learning
training examples
speech recognition
real time
named entities
learning algorithm
feature selection
object tracking
low dimensional
appearance model
small number
cross lingual
active learning
data sets
decision trees