THUMT: An Open Source Toolkit for Neural Machine Translation
This provides an incremental improvement for researchers and practitioners in natural language processing by offering a new open-source toolkit with enhanced features and performance.
The paper tackles the need for an open-source toolkit for neural machine translation by introducing THUMT, which implements attention-based encoder-decoder frameworks and supports multiple training criteria, and experiments on Chinese-English datasets show that THUMT using minimum risk training significantly outperforms the state-of-the-art toolkit GroundHog.
This paper introduces THUMT, an open-source toolkit for neural machine translation (NMT) developed by the Natural Language Processing Group at Tsinghua University. THUMT implements the standard attention-based encoder-decoder framework on top of Theano and supports three training criteria: maximum likelihood estimation, minimum risk training, and semi-supervised training. It features a visualization tool for displaying the relevance between hidden states in neural networks and contextual words, which helps to analyze the internal workings of NMT. Experiments on Chinese-English datasets show that THUMT using minimum risk training significantly outperforms GroundHog, a state-of-the-art toolkit for NMT.