FactPEGASUS: Factuality-Aware Pre-training and Fine-tuning for Abstractive Summarization

By David Wan, Mohit Bansal. 2022

78 citations recorded by the source corpus.

We present FactPEGASUS, an abstractive summarization model that addresses the problem of factuality during pre-training and fine-tuning: (1) We augment the sentence selection strategy of PEGASUS’s (Zhang et al., 2019) pre-training objective to create pseudo-summaries that are both important and factual; (2) We introduce three complementary components for fine-tuning. The corrector removes hallucinations present in the reference summary, the contrastor uses contrastive learning to better differentiate nonfactual summaries from factual ones, and the connector bridges the gap between the pre-training and fine-tuning for better transfer of knowledge. Experiments on three downstream tasks demonstrate that FactPEGASUS substantially improves factuality evaluated by multiple automatic metrics and humans. Our thorough analysis suggests that FactPEGASUS is more factual than using the original pre-training objective in zero-shot and few-shot settings, retains factual behavior more robustly than strong baselines, and does not rely entirely on becoming more extractive to improve factuality.

Subfields: Safety, Robustness & Interpretability, Summarization

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Safety, Robustness & Interpretability

1. Vidhisha Balachandran
2. Yulia Tsvetkov
3. Hannaneh Hajishirzi
4. William W. Cohen

In Summarization

1. Vidhisha Balachandran
2. Yulia Tsvetkov
3. Hannaneh Hajishirzi
4. William W. Cohen

Citations in the corpus

• Correcting Diverse Factual Errors in Abstractive Summarization via Post-Editing and Language Model Infilling (2022, methodology): “It is essential to use factually detectors to ensure summaries are factual before real world usage of any corrected summary.”