Neural Lattice Language Models

TL;DR

This paper introduces neural lattice language models that incorporate linguistic structures like polysemy and multi-word expressions, improving language modeling performance across English and Chinese tasks.

Contribution

It presents a novel lattice-based framework for language modeling that integrates linguistic intuitions and demonstrates significant perplexity improvements.

Findings

English models improve perplexity by 9.95%

Chinese models improve perplexity by 20.94%

Lattice models outperform baseline models

Abstract

In this work, we propose a new language modeling paradigm that has the ability to perform both prediction and moderation of information flow at multiple granularities: neural lattice language models. These models construct a lattice of possible paths through a sentence and marginalize across this lattice to calculate sequence probabilities or optimize parameters. This approach allows us to seamlessly incorporate linguistic intuitions - including polysemy and existence of multi-word lexical items - into our language model. Experiments on multiple language modeling tasks show that English neural lattice language models that utilize polysemous embeddings are able to improve perplexity by 9.95% relative to a word-level baseline, and that a Chinese model that handles multi-character tokens is able to improve perplexity by 20.94% relative to a character-level baseline.