This work presents a decoding architecture that fuses the information from a neural translation model and the context semantics enclosed in a semantic space language model based on word embeddings. The method extends the beam search decoding process and therefore can be applied to any neural machine translation framework. With this, we sidestep two drawbacks of current document-level systems: (ⅰ) we do not modify the training process so there is no increment in training time, and (ⅱ) we do not require document-level annotated data. We analyze the impact of the fusion system approach and its parameters on the final translation quality for English-Spanish. We obtain consistent and statistically significant improvements in terms of BLEU and METEOR and observe how the fused systems are able to handle synonyms to propose more adequate translations as well as help the system to disambiguate among several translation candidates for a word.
展开▼