References:
Use the following command to load this dataset in TFDS:
ds = tfds.load('huggingface:hover')
- Description:
HoVer is an open-domain, many-hop fact extraction and claim verification dataset built upon the Wikipedia corpus. The original 2-hop claims are adapted from question-answer pairs from HotpotQA. It is collected by a team of NLP researchers at UNC Chapel Hill and Verisk Analytics.
- License: No known license
- Version: 1.1.0
- Splits:
| Split | Examples |
|---|---|
'test' |
4000 |
'train' |
18171 |
'validation' |
4000 |
- Features:
{
"id": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"uid": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"claim": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"supporting_facts": [
{
"key": {
"dtype": "string",
"id": null,
"_type": "Value"
},
"value": {
"dtype": "int32",
"id": null,
"_type": "Value"
}
}
],
"label": {
"num_classes": 2,
"names": [
"NOT_SUPPORTED",
"SUPPORTED"
],
"names_file": null,
"id": null,
"_type": "ClassLabel"
},
"num_hops": {
"dtype": "int32",
"id": null,
"_type": "Value"
},
"hpqa_id": {
"dtype": "string",
"id": null,
"_type": "Value"
}
}