@article{zhu2021enhancing,
author = {Zhu, Chenguang and Hinthorn, William and Xu, Ruochen and Zeng, Qingkai and Zeng, Michael and Huang, Xuedong and Jiang, Meng},
title = {Enhancing Factual Consistency of Abstractive Summarization},
year = {2021},
month = {June},
abstract = {Automatic abstractive summaries are found to often distort or fabricate facts in the article. This inconsistency between summary and original text has seriously impacted its applicability. We propose a fact-aware summarization model FASum to extract and integrate factual relations into the summary generation process via graph attention. We then design a factual corrector model FC to automatically correct factual errors from summaries generated by existing systems. Empirical results show that the fact-aware summarization can produce abstractive summaries with higher factual consistency compared with existing systems, and the correction model improves the factual consistency of given summaries via modifying only a few keywords.},
url = {http://approjects.co.za/?big=en-us/research/publication/enhancing-factual-consistency-of-abstractive-summarization/},
journal = {North American Chapter of the Association for Computational Linguistics (NAACL) 2021},
}