We present a novel iterative, edit-based approach to unsupervised sentence simplification. Our model is guided by a scoring function involving fluency, simplicity, and meaning preservation. Then, we iteratively perform word and phrase-level edits on the complex sentence. Compared with previous approaches, our model does not require a parallel training set, but is more controllable and interpretable. Experiments on Newsela and WikiLarge datasets show that our approach is nearly as effective as state-of-the-art supervised approaches.
@article{arxiv.2006.09639,
title = {Iterative Edit-Based Unsupervised Sentence Simplification},
author = {Dhruv Kumar and Lili Mou and Lukasz Golab and Olga Vechtomova},
journal= {arXiv preprint arXiv:2006.09639},
year = {2020}
}