Medical Text Simplification: Optimizing for Readability with Unlikelihood Training and Reranked Beam Search Decoding

TL;DR

This paper introduces novel training and decoding techniques, including unlikelihood loss and reranked beam search, to enhance the readability and simplicity of medical text simplification models, achieving improved performance on multiple datasets.

Contribution

The paper proposes a new unlikelihood loss and reranked beam search decoding method specifically designed to improve medical text simplification.

Findings

Improved readability metrics on three datasets

Enhanced simplicity and diversity in generated medical texts

Promising results for future medical text simplification applications

Abstract

Text simplification has emerged as an increasingly useful application of AI for bridging the communication gap in specialized fields such as medicine, where the lexicon is often dominated by technical jargon and complex constructs. Despite notable progress, methods in medical simplification sometimes result in the generated text having lower quality and diversity. In this work, we explore ways to further improve the readability of text simplification in the medical domain. We propose (1) a new unlikelihood loss that encourages generation of simpler terms and (2) a reranked beam search decoding method that optimizes for simplicity, which achieve better performance on readability metrics on three datasets. This study's findings offer promising avenues for improving text simplification in the medical field.