Retrofitting Word Vectors to Semantic Lexicons

TL;DR

This paper introduces a method to enhance word vectors by integrating semantic lexicon information, significantly improving their semantic quality across multiple languages and outperforming previous techniques.

Contribution

It presents a novel, assumption-free approach to refine existing word vectors using semantic lexicons, boosting their performance on lexical semantic tasks.

Findings

Substantial improvements in semantic tasks across languages.

Outperforms prior lexicon integration methods.

Effective with various initial word vector models.

Abstract

Vector space word representations are learned from distributional information of words in large corpora. Although such statistics are semantically informative, they disregard the valuable information that is contained in semantic lexicons such as WordNet, FrameNet, and the Paraphrase Database. This paper proposes a method for refining vector space representations using relational information from semantic lexicons by encouraging linked words to have similar vector representations, and it makes no assumptions about how the input vectors were constructed. Evaluated on a battery of standard lexical semantic evaluation tasks in several languages, we obtain substantial improvements starting with a variety of word vector models. Our refinement method outperforms prior techniques for incorporating semantic lexicons into the word vector training algorithms.