Explaining Sources of Uncertainty in Automated Fact-Checking
Jingyi Sun Greta Warren Irina Shklovski Isabelle Augenstein · 2025
Read Explaining Sources of Uncertainty in Automated Fact-Checking by Jingyi Sun Greta Warren Irina Shklovski Isabelle Augenstein (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2505-17855.txt.
Description
An automated fact-checker that just says "unsure" leaves its user stuck: the doubt may come from evidence that clashes, and naming the clash would help. Jingyi Sun, Greta Warren, Irina Shklovski, and Isabelle Augenstein built a system in 2025 that explains doubt in words. It first finds the spans of claim and evidence that agree or conflict, then writes the explanation around those links.
Across three language models and two fact-checking datasets, the explanations tracked the model's true uncertainty better than plain prompting did. Human judges found them more helpful, more informative, and tighter in logic. The method needs no retraining and can wrap around any open model.
Catalog record
| CREATOR | Jingyi Sun Greta Warren Irina Shklovski Isabelle Augenstein |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 63 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2505-17855