Factual Error Correction for Abstractive Summarization Models

Meng Cao, Yue Dong, Jiapeng Wu, Jackie Chi Kit Cheung


Abstract
Neural abstractive summarization systems have achieved promising progress, thanks to the availability of large-scale datasets and models pre-trained with self-supervised methods. However, ensuring the factual consistency of the generated summaries for abstractive summarization systems is a challenge. We propose a post-editing corrector module to address this issue by identifying and correcting factual errors in generated summaries. The neural corrector model is pre-trained on artificial examples that are created by applying a series of heuristic transformations on reference summaries. These transformations are inspired by the error analysis of state-of-the-art summarization model outputs. Experimental results show that our model is able to correct factual errors in summaries generated by other neural summarization models and outperforms previous models on factual consistency evaluation on the CNN/DailyMail dataset. We also find that transferring from artificial error correction to downstream settings is still very challenging.
Anthology ID:
2020.emnlp-main.506
Volume:
Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)
Month:
November
Year:
2020
Address:
Online
Venue:
EMNLP
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
6251–6258
Language:
URL:
https://aclanthology.org/2020.emnlp-main.506
DOI:
10.18653/v1/2020.emnlp-main.506
Bibkey:
Copy Citation:
PDF:
https://aclanthology.org/2020.emnlp-main.506.pdf
Video:
 https://slideslive.com/38939120
Code
 mcao610/Factual-Error-Correction