In our lab meeting tommorow, Hassan will give us an overview of the syntax-aware NMT systems. here’s the abstract of his talk:
The idea to take advantage of syntax in Machine Translation to produce better translations has been suggested many years ago (Williams et al., 2016) and was used by many state-of-the-art Statistical Machine Translation models of the time, e.g. Hiero (Chiang, 2005), SAMT (Zollmann and Venugopal, 2006) and “GHKM rules” based models (Galley et al., 2006). Neural Machine Translation systems, however, were initially introduced without explicitly taking the syntax into account while performing the translation. Recently, Syntax-aware Neural Machine Translation systems are examining the idea to take advantage of the syntax of the sentences being translated while doing the best practices of NMT. This talk is going to perform a brief overview of the recent works done in the area of Syntax-aware Neural Machine Translation systems.
Wednesday, Feb 14, 10-11 AM, Location: TASC1 9408.