Lattice-Based Recurrent Neural Network Encoders for Neural Machine Translation

Abstract

Neural machine translation (NMT) heavily relies on word-level modelling to learn semantic representations of input sentences. However, for languages without natural word delimiters (e.g., Chinese) where input sentences have to be tokenized first, conventional NMT is confronted with two issues: 1) it is difficult to find an optimal tokenization granularity for source sentence modelling, and 2) errors in 1-best tokenizations may propagate to the encoder of NMT. To handle these issues, we propose word-lattice based Recurrent Neural Network (RNN) encoders for NMT, which generalize the standard RNN to word lattice topology. The proposed encoders take as input a word lattice that compactly encodes multiple tokenizations, and learn to generate new hidden states from arbitrarily many inputs and hidden states in preceding time steps. As such, the word-lattice based encoders not only alleviate the negative impact of tokenization errors but also are more expressive and flexible to embed input sentences. Experiment results on Chinese-English translation demonstrate the superiorities of the proposed encoders over the conventional encoder.

Cited by in corpus (16)

• PatchRNN: A Deep Learning-Based System for Security Patch Identification
• Chinese NER Using Lattice LSTM
• Confusion2Vec: Towards Enriching Vector Space Word Representations with Representational Ambiguities
• Porous Lattice-based Transformer Encoder for Chinese NER
• Deconvolution-Based Global Decoding for Neural Machine Translation
• Memorized Sparse Backpropagation
• Lattice-Based Transformer Encoder for Neural Machine Translation
• Bag-of-Words as Target for Neural Machine Translation
• Lattice Transformer for Speech Translation
• Lattice CNNs for Matching Based Chinese Question Answering
• Lexicon-constrained Copying Network for Chinese Abstractive Summarization
• Tackling Graphical NLP problems with Graph Recurrent Networks
• Neural Machine Translation: A Review and Survey
• The Helsinki Neural Machine Translation System
• Learning Domain Specific Language Models for Automatic Speech Recognition through Machine Translation
• Structural Information Preserving for Graph-to-Text Generation