Multi-Source Neural Translation

Barret Zoph, Kevin K. Knight · 2016

We build a multi-source machine translation model and train it to maximize the probability of a target English string given French and German sources.Using the neural encoderdecoder framework, we explore several combination methods and report up to +4.8 Bleu increases on top of a very strong attentionbased neural translation model.

Read the paper · More papers on PaperTik