Multilingual NMT with a Language-Independent Attention Bridge

Explore this paper's citation graph

Summary

A new framework for the efficient development of multilingual neural machine translation (NMT) using this model and scheduled training and achieves substantial improvements over strong bilingual models and performs well for zero-shot translation, which demonstrates its ability of abstraction and transfer learning.

Type
preprint
Published
2018-11-01
Cited by
46
References
36
Access
Open access

Keywords

Computer science, Natural language processing, Artificial intelligence, Bridge (graph theory), Abstraction

References

Cited by

Related papers