Improving Neural Language Models by Segmenting, Attending, and Predicting the Future

Abstract

Common language models typically predict the next word given the context. In this work, we propose a method that improves language modeling by learning to align the given context and the following phrase. The model does not require any linguistic annotation of phrase segmentation. Instead, we define syntactic heights and phrase segmentation rules, enabling the model to automatically induce phrases, recognize their task-specific heads, and generate phrase embeddings in an unsupervised learning manner. Our method can easily be applied to language models with different network architectures since an independent module is used for phrase induction and context-phrase alignment, and no change is required in the underlying language modeling network. Experiments have shown that our model outperformed several strong baseline models on different data sets. We achieved a new state-of-the-art performance of 17.4 perplexity on the Wikitext-103 dataset. Additionally, visualizing the outputs of the phrase induction module showed that our model is able to learn approximate phrase-level structural knowledge without any annotation.

Accepted by ACL 2019

References in corpus (18)

• Distributed Representations of Words and Phrases and their Compositionality
• An Empirical Evaluation of Generic Convolutional and Recurrent Networks for Sequence Modeling
• DARTS: Differentiable Architecture Search
• Character-Aware Neural Language Models
• Efficient Neural Architecture Search via Parameter Sharing
• Pointer Sentinel Mixture Models
• Regularizing and Optimizing LSTM Language Models
• Quasi-Recurrent Neural Networks
• Long Short-Term Memory-Networks for Machine Reading
• Tying Word Vectors and Word Classifiers: A Loss Framework for Language Modeling
• An Analysis of Neural Language Modeling at Multiple Scales
• Ordered Neurons: Integrating Tree Structures into Recurrent Neural Networks
• Breaking the Softmax Bottleneck: A High-Rank RNN Language Model
• Recurrent Memory Networks for Language Modeling
• Neural Language Modeling by Jointly Learning Syntax and Lexicon
• Fast Parametric Learning with Activation Memorization
• Pushing the bounds of dropout
• Straight to the Tree: Constituency Parsing with Neural Syntactic Distance