AI & MLSpecializedAIML-NLP-005
Multilingual NLP
One model, many languages: shared vocabularies, cross-lingual transfer, and the costs borne by low-resource languages.
- Difficulty
- 5 / 5
- Importance
- 3 / 5
- Study time
- 1–2 h
- Kind of work
- Mixed
Why it matters
The learner can explain how tokenizer fertility and data imbalance disadvantage low-resource languages in a multilingual model.