# Word Usage Similarity Estimation with Sentence Representations and Automatic Substitutes

## TL;DR

This paper introduces supervised models utilizing contextualized embeddings and lexical substitutes to estimate word usage similarity across contexts, outperforming previous methods on benchmark datasets.

## Contribution

It presents a novel approach combining contextualized embeddings with lexical substitute annotations for improved usage similarity estimation.

## Key findings

- Supervised models outperform previous methods in similarity tasks.
- Contextualized embeddings like BERT and ELMo enhance prediction accuracy.
- Lexical substitute annotations improve model performance.

## Abstract

Usage similarity estimation addresses the semantic proximity of word instances in different contexts. We apply contextualized (ELMo and BERT) word and sentence embeddings to this task, and propose supervised models that leverage these representations for prediction. Our models are further assisted by lexical substitute annotations automatically assigned to word instances by context2vec, a neural model that relies on a bidirectional LSTM. We perform an extensive comparison of existing word and sentence representations on benchmark datasets addressing both graded and binary similarity. The best performing models outperform previous methods in both settings.