Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing

Research output: Contribution to journal › Journal article › Research › peer-review

Standard

Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing. / Jolly, Shailza; Atanasova, Pepa; Augenstein, Isabelle.

In: Information (Switzerland), Vol. 13, No. 10, 500, 2022, p. 1-18.

Research output: Contribution to journal › Journal article › Research › peer-review

Harvard

Jolly, S, Atanasova, P & Augenstein, I 2022, 'Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing', Information (Switzerland), vol. 13, no. 10, 500, pp. 1-18. https://doi.org/10.3390/info13100500

APA

Jolly, S., Atanasova, P., & Augenstein, I. (2022). Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing. Information (Switzerland), 13(10), 1-18. [500]. https://doi.org/10.3390/info13100500

Vancouver

Jolly S, Atanasova P, Augenstein I. Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing. Information (Switzerland). 2022;13(10):1-18. 500. https://doi.org/10.3390/info13100500

Author

Jolly, Shailza ; Atanasova, Pepa ; Augenstein, Isabelle. / Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing. In: Information (Switzerland). 2022 ; Vol. 13, No. 10. pp. 1-18.

Bibtex

@article{2d787a09efa7436aa115407c7b77ed27,

title = "Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing",

abstract = "Fact-checking systems have become important tools to verify fake and misguiding news. These systems become more trustworthy when human-readable explanations accompany the veracity labels. However, manual collection of these explanations is expensive and time-consuming. Recent work has used extractive summarization to select a sufficient subset of the most important facts from the ruling comments (RCs) of a professional journalist to obtain fact-checking explanations. However, these explanations lack fluency and sentence coherence. In this work, we present an iterative edit-based algorithm that uses only phrase-level edits to perform unsupervised post-editing of disconnected RCs. To regulate our editing algorithm, we use a scoring function with components including fluency and semantic preservation. In addition, we show the applicability of our approach in a completely unsupervised setting. We experiment with two benchmark datasets, namely LIAR-PLUS and PubHealth. We show that our model generates explanations that are fluent, readable, non-redundant, and cover important information for the fact check.",

keywords = "explainable AI, fact-checking, natural language generation",

author = "Shailza Jolly and Pepa Atanasova and Isabelle Augenstein",

note = "Publisher Copyright: {\textcopyright} 2022 by the authors.",

year = "2022",

doi = "10.3390/info13100500",

language = "English",

volume = "13",

pages = "1--18",

journal = "Information (Switzerland)",

issn = "2078-2489",

publisher = "MDPI - Open Access Publishing",

number = "10",

}

RIS

TY - JOUR

T1 - Generating Fluent Fact Checking Explanations with Unsupervised Post-Editing

AU - Jolly, Shailza

AU - Atanasova, Pepa

AU - Augenstein, Isabelle

PY - 2022

Y1 - 2022

N2 - Fact-checking systems have become important tools to verify fake and misguiding news. These systems become more trustworthy when human-readable explanations accompany the veracity labels. However, manual collection of these explanations is expensive and time-consuming. Recent work has used extractive summarization to select a sufficient subset of the most important facts from the ruling comments (RCs) of a professional journalist to obtain fact-checking explanations. However, these explanations lack fluency and sentence coherence. In this work, we present an iterative edit-based algorithm that uses only phrase-level edits to perform unsupervised post-editing of disconnected RCs. To regulate our editing algorithm, we use a scoring function with components including fluency and semantic preservation. In addition, we show the applicability of our approach in a completely unsupervised setting. We experiment with two benchmark datasets, namely LIAR-PLUS and PubHealth. We show that our model generates explanations that are fluent, readable, non-redundant, and cover important information for the fact check.

AB - Fact-checking systems have become important tools to verify fake and misguiding news. These systems become more trustworthy when human-readable explanations accompany the veracity labels. However, manual collection of these explanations is expensive and time-consuming. Recent work has used extractive summarization to select a sufficient subset of the most important facts from the ruling comments (RCs) of a professional journalist to obtain fact-checking explanations. However, these explanations lack fluency and sentence coherence. In this work, we present an iterative edit-based algorithm that uses only phrase-level edits to perform unsupervised post-editing of disconnected RCs. To regulate our editing algorithm, we use a scoring function with components including fluency and semantic preservation. In addition, we show the applicability of our approach in a completely unsupervised setting. We experiment with two benchmark datasets, namely LIAR-PLUS and PubHealth. We show that our model generates explanations that are fluent, readable, non-redundant, and cover important information for the fact check.

KW - explainable AI

KW - fact-checking

KW - natural language generation

UR - http://www.scopus.com/inward/record.url?scp=85140652386&partnerID=8YFLogxK

U2 - 10.3390/info13100500

DO - 10.3390/info13100500

M3 - Journal article

AN - SCOPUS:85140652386

VL - 13

SP - 1

EP - 18

JO - Information (Switzerland)

JF - Information (Switzerland)

SN - 2078-2489

IS - 10

M1 - 500

ER -

ID: 324680448

Department of Computer Science