A character-level neural translation model reconstructs Latin proto-words from Romance cognates with 64 percent exact accuracy and learns sound-change rules consistent with historical linguistics.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CL 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Ab Antiquo: Neural Proto-language Reconstruction
A character-level neural translation model reconstructs Latin proto-words from Romance cognates with 64 percent exact accuracy and learns sound-change rules consistent with historical linguistics.