Exploiting Syntactic Structure for Better Language Modeling: A Syntactic Distance Approach
arXiv:2005.05864
Abstract
It is commonly believed that knowledge of syntactic structure should improve language modeling. However, effectively and computationally efficiently incorporating syntactic structure into neural language models has been a challenging topic. In this paper, we make use of a multi-task objective, i.e., the models simultaneously predict words as well as ground truth parse trees in a form called "syntactic distances", where information between these two separate objectives shares the same intermediate representation. Experimental results on the Penn Treebank and Chinese Treebank datasets show that when ground truth parse trees are provided as additional training signals, the model is able to achieve lower perplexity and induce trees with better quality.
ACL20
References in corpus (6)
- Neural Architecture Search with Reinforcement Learning
- Pointer Sentinel Mixture Models
- Regularizing and Optimizing LSTM Language Models
- A Minimal Span-Based Neural Constituency Parser
- A Generative Parser with a Discriminative Recognition Algorithm
- Scalable Syntax-Aware Language Models Using Knowledge Distillation