the
∞
.ai
Search
/
Concepts
trail ·
0 nodes
Every concept
Multilingual
8 concepts, 8 reviewed and merged.
Data
Language Imbalance
Web text is overwhelmingly a handful of languages.
verified
Foundations
Multilingual Model
One model for many languages, sharing parameters across all of them.
verified
Interface
Script and Orthography
Writing systems differ in ways that reach all the way down to the byte level.
verified
Tokenizer Fairness
A tokenizer trained mostly on English gives English the shortest encodings.
verified
Methods
Cross-Lingual Transfer
Fine-tune a multilingual model on a task in English and it often performs that task in languages it never saw…
verified
Machine Translation
The task that produced the encoder-decoder, attention, and eventually the transformer.
verified
Regimes
Code-Switching
People switch languages mid-sentence, and they do it by rules rather than at random.
verified
Low-Resource Language
Most of the world's languages have almost no digital text.
verified