EMNLP 2023short main0 citations
Simplicity Level Estimate (SLE): A Learned Reference-Less Metric for Sentence Simplification
Liam Cripwell, Joël Legrand, Claire Gardent
Abstract
Automatic evaluation for sentence simplification remains a challenging problem. Most popular evaluation metrics require multiple high-quality references -- something not readily available for simplification -- which makes it difficult to test performance on unseen domains. Furthermore, most existing metrics conflate simplicity with correlated attributes such as fluency or meaning preservation. We propose a new learned evaluation metric --- SLE --- which focuses on simplicity, outperforming almost all existing metrics in terms of correlation with human judgements.
simplificationevaluationquality estimation
BibTeX
@inproceedings{
cripwell2023simplicity,
title={Simplicity Level Estimate ({SLE}): A Learned Reference-Less Metric for Sentence Simplification},
author={Liam Cripwell and Jo{\"e}l Legrand and Claire Gardent},
booktitle={The 2023 Conference on Empirical Methods in Natural Language Processing},
year={2023},
url={https://openreview.net/forum?id=Gp8EmdJLUj}
}