Knowledge

NMT 95 Constrained Parameter Initialisation for Deep Transformers in Neural MT

Author: Dr. Patrik Lambert, Senior Machine Translation Scientist @ Iconic Introduction As the Transformer model is the state of the art in Neural MT, researchers have tried to build wider (with higher dimension vectors) and deeper (with more layers) Transformer networks. Wider networks are more costly in terms of training and generation time, thus they are not the best option in production environments. However, adding encoder layers...

Read More
NMT 94 Unsupervised Parallel Sentence Extraction with Parallel Segment Detection Helps Machine Translation

Author: Dr. Chao-Hong Liu, Machine Translation Scientist @ Iconic Introduction Curating corpora of quality sentence pairs is a fundamental task to building Machine Translation (MT) systems. This resource can be availed from Translation Memory (TM) systems where the human translations are recorded. However, in most cases we don’t have TM databases but comparable corpora, e.g. news articles of the same story in different languages. In this post,...

Read More
NMT 93 Semantic Neural Machine Translation using AMR

Author: Dr. Karin Sim, Machine Translation Scientist @ Iconic Introduction Semantic representations were part of the very early Machine Translation (MT) systems, yet have had little role in recent Neural MT (NMT) systems. Given that a good translation should reflect the meaning of the source text, this seems an important area to focus on, particularly since the abstraction could potentially help handle data sparsity. In today’s blog...

Read More
NMT 92 The Importance of References in Evaluating MT Output

Author: Dr. Carla Parra Escartín, Global Program Manager @ Iconic Introduction Over the years, BLEU has become the “de facto standard” for Machine Translation automatic evaluation. However, and despite being the metric being referenced in all MT research papers, it is equally criticized for not providing a reliable evaluation of the MT output. In today’s blog post we look at the work done by Freitag et al....

Read More
NMT Podcast Series

Following the popularity of our latest Webinar series entitled Neural Machine Translation: The Past - The Present - The Future, we are pleased to announce that it is now available via podcast. The interest and feedback from the series was phenomenal, which prompted us to also offer it as a podcast. The three podcasts will cover the history of Neural MT, albeit a relatively short...

Read More