Figure 3 - available via license: Creative Commons Attribution 4.0 International
Content may be subject to copyright.
Source publication
To what extent can neural network models learn generalizations about language structure, and how do we find out what they have learned? We explore these questions by training neural models for a range of natural language processing tasks on a massively multilingual dataset of Bible translations in 1295 languages. The learned language representation...
Contexts in source publication
Context 1
... Order of numeral and noun, for the WordLM representations (Figure 3). Note that no representations obtained a mean F 1 above 0.7 when trained on URIEL data. ...