Syntactically Guided Neural Machine Translation

Abstract

We investigate the use of hierarchical phrase-based SMT lattices in end-to-end neural machine translation (NMT). Weight pushing transforms the Hiero scores for complete translation hypotheses, with the full translation grammar score and full n-gram language model score, into posteriors compatible with NMT predictive probabilities. With a slightly modified NMT beam-search decoder we find gains over both Hiero and NMT decoding alone, with practical advantages in extending NMT to very large input and output vocabularies.

ACL 2016

References in corpus (6)

• Neural Machine Translation by Jointly Learning to Align and Translate
• Sequence to Sequence Learning with Neural Networks
• Theano: new features and speed improvements
• Modeling Coverage for Neural Machine Translation
• Blocks and Fuel: Frameworks for deep learning
• Neural Reranking Improves Subjective Quality of Machine Translation: NAIST at WAT2015

Cited by in corpus (22)

• Semantic Neural Machine Translation using AMR
• Neural Machine Translation with Reconstruction
• Syntactically Supervised Transformers for Faster Neural Machine Translation
• Learning to Parse and Translate Improves Neural Machine Translation
• Memory-augmented Neural Machine Translation
• Translating Phrases in Neural Machine Translation
• Towards Bidirectional Hierarchical Representations for Attention-Based Neural Machine Translation
• Neural Machine Translation Advised by Statistical Machine Translation
• Improving Neural Machine Translation through Phrase-based Forced Decoding
• Why not be Versatile? Applications of the SGNMT Decoder for Machine Translation
• Recurrent Graph Syntax Encoder for Neural Machine Translation
• Merging External Bilingual Pairs into Neural Machine Translation
• On the Integration of LinguisticFeatures into Statistical and Neural Machine Translation
• Automatic Post-Editing for Machine Translation
• Top-down Tree Structured Decoding with Syntactic Connections for Neural Machine Translation and Parsing
• Refining Source Representations with Relation Networks for Neural Machine Translation
• Neural Machine Translation with Source-Side Latent Graph Parsing
• Discrete Structural Planning for Neural Machine Translation
• Phrase Table as Recommendation Memory for Neural Machine Translation
• Tackling Graphical NLP problems with Graph Recurrent Networks
• Neural Machine Translation with Explicit Phrase Alignment
• Learning synchronous context-free grammars with multiple specialised non-terminals for hierarchical phrase-based translation