Compositional Morphology for Word Representations and Language Modelling

TL;DR

This paper introduces a scalable compositional morphological approach integrated into probabilistic language models, improving word similarity and translation performance across multiple languages.

Contribution

It presents a novel, efficient method for incorporating morphological information into language models, enhancing translation quality and word similarity tasks.

Findings

Improved word similarity performance

Reduced perplexity in language modeling

Up to 1.2 BLEU point improvement in translation

Abstract

This paper presents a scalable method for integrating compositional morphological representations into a vector-based probabilistic language model. Our approach is evaluated in the context of log-bilinear language models, rendered suitably efficient for implementation inside a machine translation decoder by factoring the vocabulary. We perform both intrinsic and extrinsic evaluations, presenting results on a range of languages which demonstrate that our model learns morphological representations that both perform well on word similarity tasks and lead to substantial reductions in perplexity. When used for translation into morphologically rich languages with large vocabularies, our models obtain improvements of up to 1.2 BLEU points relative to a baseline system using back-off n-gram models.