Back-translation based approaches have recently lead to significant progress in unsupervised sequence-to-sequence tasks such as machine translation or style transfer. In this work, we extend the paradigm to the problem of learning a sentence summarization system from unaligned data. We present several initial models which rely on the asymmetrical nature of the task to perform the first back-translation step, and demonstrate the value of combining the data created by these diverse initialization methods. Our system outperforms the current state-of-the-art for unsupervised sentence summarization from fully unaligned data by over 2 ROUGE, and matches the performance of recent semi-supervised approaches.
@article{arxiv.1908.08566,
title = {Unsupervised Text Summarization via Mixed Model Back-Translation},
author = {Yacine Jernite},
journal= {arXiv preprint arXiv:1908.08566},
year = {2019}
}