New answers tagged word-embedding
1
vote
Accepted
Is having many tokens generally good for NLP performance?
Indeed several studies provide evidence that the design of tokenizers significantly impacts transformer model performance such as the original BERT paper that introduced WordPiece tokenization ...
Top 50 recent answers are included
Related Tags
word-embedding × 98natural-language-processing × 57
word2vec × 20
transformer × 17
embeddings × 14
neural-networks × 12
recurrent-neural-networks × 9
deep-learning × 8
machine-learning × 7
attention × 7
bert × 7
terminology × 6
text-classification × 6
long-short-term-memory × 5
comparison × 4
reference-request × 4
pytorch × 3
data-preprocessing × 3
books × 3
pretrained-models × 3
glove × 3
large-language-models × 3
classification × 2
tensorflow × 2
keras × 2