Falsesum: Generating Document-level NLI Examples for Recognizing Factual Inconsistency in Summarization

Prasetya Utama, Joshua Bambrick, Nafise Moosavi, Iryna Gurevych


Abstract
Neural abstractive summarization models are prone to generate summaries that are factually inconsistent with their source documents. Previous work has introduced the task of recognizing such factual inconsistency as a downstream application of natural language inference (NLI). However, state-of-the-art NLI models perform poorly in this context due to their inability to generalize to the target task. In this work, we show that NLI models can be effective for this task when the training data is augmented with high-quality task-oriented examples. We introduce Falsesum, a data generation pipeline leveraging a controllable text generation model to perturb human-annotated summaries, introducing varying types of factual inconsistencies. Unlike previously introduced document-level NLI datasets, our generated dataset contains examples that are diverse and inconsistent yet plausible. We show that models trained on a Falsesum-augmented NLI dataset improve the state-of-the-art performance across four benchmarks for detecting factual inconsistency in summarization.
Anthology ID:
2022.naacl-main.199
Volume:
Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies
Month:
July
Year:
2022
Address:
Seattle, United States
Editors:
Marine Carpuat, Marie-Catherine de Marneffe, Ivan Vladimir Meza Ruiz
Venue:
NAACL
SIG:
Publisher:
Association for Computational Linguistics
Note:
Pages:
2763–2776
Language:
URL:
https://aclanthology.org/2022.naacl-main.199
DOI:
10.18653/v1/2022.naacl-main.199
Bibkey:
Cite (ACL):
Prasetya Utama, Joshua Bambrick, Nafise Moosavi, and Iryna Gurevych. 2022. Falsesum: Generating Document-level NLI Examples for Recognizing Factual Inconsistency in Summarization. In Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies, pages 2763–2776, Seattle, United States. Association for Computational Linguistics.
Cite (Informal):
Falsesum: Generating Document-level NLI Examples for Recognizing Factual Inconsistency in Summarization (Utama et al., NAACL 2022)
Copy Citation:
PDF:
https://aclanthology.org/2022.naacl-main.199.pdf
Video:
 https://aclanthology.org/2022.naacl-main.199.mp4
Code
 joshbambrick/falsesum
Data
CNN/Daily MailMultiNLISummEval