FEVEROUS: Fact Extraction and VERification Over Unstructured and Structured information
This addresses the problem of detecting misinformation by incorporating structured data like tables, which is incremental as it extends existing text-only benchmarks.
The authors tackled fact verification by creating a new dataset, FEVEROUS, with 87,026 claims that require evidence from both text and tables, and developed a baseline model that correctly predicts evidence and verdict for 18% of claims.
Fact verification has attracted a lot of attention in the machine learning and natural language processing communities, as it is one of the key methods for detecting misinformation. Existing large-scale benchmarks for this task have focused mostly on textual sources, i.e. unstructured information, and thus ignored the wealth of information available in structured formats, such as tables. In this paper we introduce a novel dataset and benchmark, Fact Extraction and VERification Over Unstructured and Structured information (FEVEROUS), which consists of 87,026 verified claims. Each claim is annotated with evidence in the form of sentences and/or cells from tables in Wikipedia, as well as a label indicating whether this evidence supports, refutes, or does not provide enough information to reach a verdict. Furthermore, we detail our efforts to track and minimize the biases present in the dataset and could be exploited by models, e.g. being able to predict the label without using evidence. Finally, we develop a baseline for verifying claims against text and tables which predicts both the correct evidence and verdict for 18% of the claims.