Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
parquet
Sub-tasks:
fact-checking
Languages:
English
Size:
100K - 1M
License:
metadata
annotations_creators:
- crowdsourced
language_creators:
- crowdsourced
language:
- en
license:
- cc-by-sa-3.0
- gpl-3.0
multilinguality:
- monolingual
paperswithcode_id: fever
pretty_name: fever
size_categories:
- 100K<n<1M
source_datasets:
- extended|fever
task_categories:
- text-classification
task_ids:
- fact-checking
Dataset Summary
This dataset is extracted from Fever dataset (https://fever.ai), pre-processed and ready to train and evaluate. The training objective is a text classification task - given a claim and evidence, predict if evidence is related to claim.