Иерархический Машинный Перевод

State of the art Tree Structures Prediction techniques rely on bottom-up decoding. These approaches allow the use of context-free fea- tures and bottom-up features. We discuss the limitations of mainstream …...

Despite being closely related languages, German and English are characterized by important word order differences. Long- range reordering of verbs, in particular, represents a real challenge for state-of-the- art SMT …...

AbstractThe idea of two-step machine translation was introduced to divide the complexity of the search space into two independent steps: (1) lexical translation and reordering, and (2) conjugation and declination …...

This paper proposes a method for hierar- chical phrase-based stream decoding. A stream decoder is able to take a contin- uous stream of tokens as input, and seg- ments this …...