Where is Your Evidence: Improving Fact-checking by Justification Modeling

By Tariq Alhindi, Savvas Petridis, Smaranda Muresan. 2018

180 citations recorded by the source corpus.

Fact-checking is a journalistic practice that compares a claim made publicly against trusted sources of facts. Wang (2017) introduced a large dataset of validated claims from the POLITIFACT.com website (LIAR dataset), enabling the development of machine learning approaches for fact-checking. However, approaches based on this dataset have focused primarily on modeling the claim and speaker-related metadata, without considering the evidence used by humans in labeling the claims. We extend the LIAR dataset by automatically extracting the justification from the fact-checking article used by humans to label a given claim. We show that modeling the extracted justification in conjunction with the claim (and metadata) provides a significant improvement regardless of the machine learning model used (feature-based or deep learning) both in a binary classification task (true, false) and in a six-way classification task (pants on fire, false, mostly false, half true, mostly true, true).

Subfields: Artificial Intelligence, Sociology and Political Science, Natural Language Processing

How researchers see this paper

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

In Artificial Intelligence

In Sociology and Political Science

In Natural Language Processing

Citations in the corpus

Avoiding catastrophic forgetting in mitigating model biases in sentence-pair classification with elastic weight consolidation (2020, methodology): “While LiarPlus (Alhindi et al., 2018) and MultiFC (Augenstein et al., 2019) both have a similar task signature with claims and additional information provided as text, the text is not evidential (i.e. not containing a premise that would support or refute the claim).”
Few-Shot Self-Rationalization with Natural Language Prompts (2021, background): “We thus exclude more complex LIAR-PLUS (Alhindi et al., 2018) and PUBHEALTH (Kotonya and Toni, 2020), and solving algebraic word problems (Ling et al.”
Towards Debiasing Fact Verification Models (2019, background): “through common sense knowledge), contextual verification is key for a fact-checking task (Alhindi et al., 2018).”
Teach Me to Explain: A Review of Datasets for Explainable NLP (2021, background): “, 2018) natural language inference crowd ∼569K (1 or 3) LIAR-PLUS (Alhindi et al., 2018) verifying claims from text auto 12,836 COS-E V1.”
Automated Fact-Checking for Assisting Human Fact-Checkers (2021, unknown): “This has been shown to be an effective approach when assessing claim veracity [Alhindi et al., 2018].”
Using Interactive Feedback to Improve the Accuracy and Explainability of Question Answering Systems Post-Deployment (2022, methodology): “We thus collect questions for individual passages beforehand similar to Stage 1 and use these as in-148 Commonsense QA Free-form LIAR-PLUS (Alhindi et al., 2018) Fact checking Free-form QED (Lamm et al., 2021) Reading comprehension Structured NExT (Wang et al., 2019) Text classiﬁcation Structured…”
FEVEROUS: Fact Extraction and VERification Over Unstructured and Structured information (2021, unknown): “…[Popat et al., 2016, Baly et al., 2018, Augenstein et al., 2019], or bypass the retrieval challenge entirely by extracting evidence directly from the fact checking articles [Alhindi et al., 2018, Hanselowski et al., 2019, Kotonya and Toni, 2020b] or scientiﬁc literature [Wadden et al., 2020].”
On the Diversity and Limits of Human Explanations (2021, background): “Reference Task Questions and guidelines Ling et al. (2017) algebraic prob- lems cannot infer the exact question Alhindi et al. (2018) fact verification we cannot infer the exact question automatically extracting for each claim the justification that humans have provided in the fact-checking article…”
Explainable Automated Fact-Checking: A Survey (2020, background): “(2018) 23,921 PolitiFact, GossipCop 2 X Alhindi et al. (2018) 12,836 PolitiFact 6 X Thorne et al.”
Teach Me to Explain: A Review of Datasets for Explainable Natural Language Processing (2021, background): “One exception is LIAR-PLUS [4], which111 contains the conclusion paragraphs of web-scraped human-written fact-checking summaries.112 Structured Explanations (Table 5) Structured explanations take on dataset-specific forms.”
Generating Fact Checking Explanations (2020, methodology): “Even given a gold explanation (Alhindi et al. (2018) and Veracity@Just ), the macro F1 remains below 0.5.”
When retrieval outperforms generation: Dense evidence retrieval for scalable fake news detection (2025, unknown): “LIAR-RAW is an expanded version of the LIAR-PLUS dataset (Alhindi et al., 2018).”