Extractive summarization models require sentence-level labels, which are usually created heuristically (e.g., with rule-based methods) given that most summarization datasets only have document-summary pairs. Since these labels might be suboptimal, we propose a latent variable extractive model where sentences are viewed as latent variables and sentences with activated variables are used to infer gold summaries. During training the loss comes \emph{directly} from gold summaries. Experiments on the CNN/Dailymail dataset show that our model improves over a strong extractive baseline trained on heuristically approximated labels and also performs competitively to several recent models.
@article{arxiv.1808.07187,
title = {Neural Latent Extractive Document Summarization},
author = {Xingxing Zhang and Mirella Lapata and Furu Wei and Ming Zhou},
journal= {arXiv preprint arXiv:1808.07187},
year = {2018}
}