Can summarization approximate simplification? A gold standard comparison

TL;DR

This paper investigates how closely abstractive summarization can mimic text simplification by comparing BART-based summarization outputs with manual simplifications on the Newsela corpus, revealing overlaps and differences.

Contribution

It introduces a systematic comparison between summarization and simplification outputs using BART-based models and provides insights into their convergence and divergence.

Findings

ROUGE-L score of 0.654 for summarization compared to simplification

Identifies key areas of overlap between summarization and simplification outputs

Highlights differences in cohesion and content preservation

Abstract

This study explores the overlap between text summarization and simplification outputs. While summarization evaluation methods are streamlined, simplification lacks cohesion, prompting the question: how closely can abstractive summarization resemble gold-standard simplification? We address this by applying two BART-based BRIO summarization methods to the Newsela corpus, comparing outputs with manually annotated simplifications and achieving a top ROUGE-L score of 0.654. This provides insight into where summarization and simplification outputs converge and differ.