Predicting Word Similarity in Context with Referential Translation Machines

TL;DR

This paper introduces a novel approach using referential translation machines to predict word similarity in context, achieving top results in the GWSC task by framing it as a machine translation performance prediction problem.

Contribution

The paper presents a new method employing RTMs and stacked machine learning models to improve contextual word similarity prediction performance.

Findings

RTMs achieve top results in GWSC task

Framing similarity prediction as machine translation performance prediction

Stacked machine learning models enhance accuracy

Abstract

We identify the similarity between two words in English by casting the task as machine translation performance prediction (MTPP) between the words given the context and the distance between their similarities. We use referential translation machines (RTMs), which allows a common representation for training and test sets and stacked machine learning models. RTMs can achieve the top results in Graded Word Similarity in Context (GWSC) task.