We present ToTTo, an open-domain English table-to-text dataset with over 120,000 training examples that proposes a controlled generation task: given a Wikipedia table and a set of highlighted table cells, produce a one-sentence description. To obtain generated targets that are natural but also faithful to the source table, we introduce a dataset construction process where annotators directly revise existing candidate sentences from Wikipedia. We present systematic analyses of our dataset and annotation process as well as results achieved by several state-of-the-art baselines. While usually fluent, existing methods often hallucinate phrases that are not supported by the table, suggesting that this dataset can serve as a useful research benchmark for high-precision conditional text generation.
@article{arxiv.2004.14373,
title = {ToTTo: A Controlled Table-To-Text Generation Dataset},
author = {Ankur P. Parikh and Xuezhi Wang and Sebastian Gehrmann and Manaal Faruqui and Bhuwan Dhingra and Diyi Yang and Dipanjan Das},
journal= {arXiv preprint arXiv:2004.14373},
year = {2020}
}