The advent of large pre-trained language models has made it possible to make high-quality predictions on how to add or change a sentence in a document. However, the high branching factor inherent to text generation impedes the ability of even the strongest language models to offer useful editing suggestions at a more global or document level. We introduce a new task, document sketching, which involves generating entire draft documents for the writer to review and revise. These drafts are built from sets of documents that overlap in form - sharing large segments of potentially reusable text - while diverging in content. To support this task, we introduce a Wikipedia-based dataset of analogous documents and investigate the application of weakly supervised methods, including use of a transformer-based mixture of experts, together with reinforcement learning. We report experiments using automated and human evaluation methods and discuss relative merits of these models.
@article{arxiv.2106.07192,
title = {Automatic Document Sketching: Generating Drafts from Analogous Texts},
author = {Zeqiu Wu and Michel Galley and Chris Brockett and Yizhe Zhang and Bill Dolan},
journal= {arXiv preprint arXiv:2106.07192},
year = {2021}
}