SWiPE: A Dataset for Document-Level Simplification of Wikipedia Pages

TL;DR

The paper introduces SWiPE, a comprehensive dataset capturing document-level edits in Wikipedia simplification, along with models to automatically label these edits, advancing understanding of the simplification process.

Contribution

It provides a new dataset with detailed annotations of document-level edits in Wikipedia, and develops models to automatically identify and categorize these edits.

Findings

SWiPE dataset includes 5,000 annotated document pairs with over 40,000 labeled edits.

Automatic labeling models achieved up to 70.6 F-1 score, demonstrating task tractability.

Models trained on SWiPE produce more complex and fewer unwanted edits.

Abstract

Text simplification research has mostly focused on sentence-level simplification, even though many desirable edits - such as adding relevant background information or reordering content - may require document-level context. Prior work has also predominantly framed simplification as a single-step, input-to-output task, only implicitly modeling the fine-grained, span-level edits that elucidate the simplification process. To address both gaps, we introduce the SWiPE dataset, which reconstructs the document-level editing process from English Wikipedia (EW) articles to paired Simple Wikipedia (SEW) articles. In contrast to prior work, SWiPE leverages the entire revision history when pairing pages in order to better identify simplification edits. We work with Wikipedia editors to annotate 5,000 EW-SEW document pairs, labeling more than 40,000 edits with proposed 19 categories. To scale our efforts, we propose several models to automatically label edits, achieving an F-1 score of up to 70.6, indicating that this is a tractable but challenging NLU task. Finally, we categorize the edits produced by several simplification models and find that SWiPE-trained models generate more complex edits while reducing unwanted edits.