Update README.md
Browse filesUpdated documentation.
README.md
CHANGED
@@ -1,3 +1,134 @@
|
|
1 |
---
|
2 |
-
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
1 |
---
|
2 |
+
language:
|
3 |
+
- en
|
4 |
+
license:
|
5 |
+
- cc-by-nc-2.0
|
6 |
---
|
7 |
+
# Dataset Card for "scifact"
|
8 |
+
|
9 |
+
## Table of Contents
|
10 |
+
- [Dataset Description](#dataset-description)
|
11 |
+
- [Dataset Summary](#dataset-summary)
|
12 |
+
- [Dataset Structure](#dataset-structure)
|
13 |
+
- [Data Instances](#data-instances)
|
14 |
+
- [Data Fields](#data-fields)
|
15 |
+
- [Data Splits](#data-splits)
|
16 |
+
- [Additional Information](#additional-information)
|
17 |
+
- [Licensing Information](#licensing-information)
|
18 |
+
- [Citation Information](#citation-information)
|
19 |
+
|
20 |
+
## Dataset Description
|
21 |
+
|
22 |
+
- **Homepage:** [https://scifact.apps.allenai.org/](https://scifact.apps.allenai.org/)
|
23 |
+
- **Paper:** [Fact or Fiction: Verifying Scientific Claims](https://aclanthology.org/2020.emnlp-main.609/)
|
24 |
+
|
25 |
+
### Dataset Summary
|
26 |
+
|
27 |
+
SciFact.
|
28 |
+
|
29 |
+
This is a dataset of expert-written scientific claims paired with evidence-containing abstracts and annotated with labels and rationales.
|
30 |
+
|
31 |
+
## Dataset Structure
|
32 |
+
|
33 |
+
### Data Instances
|
34 |
+
|
35 |
+
#### claims
|
36 |
+
|
37 |
+
- **Size of downloaded dataset files:** 2.72 MB
|
38 |
+
- **Size of the generated dataset:** 0.25 MB
|
39 |
+
- **Total amount of disk used:** 2.97 MB
|
40 |
+
|
41 |
+
An example of 'validation' looks as follows.
|
42 |
+
```
|
43 |
+
{
|
44 |
+
"cited_doc_ids": [14717500],
|
45 |
+
"claim": "1,000 genomes project enables mapping of genetic sequence variation consisting of rare variants with larger penetrance effects than common variants.",
|
46 |
+
"evidence_doc_id": "14717500",
|
47 |
+
"evidence_label": "SUPPORT",
|
48 |
+
"evidence_sentences": [2, 5],
|
49 |
+
"id": 3
|
50 |
+
}
|
51 |
+
```
|
52 |
+
|
53 |
+
#### corpus
|
54 |
+
|
55 |
+
- **Size of downloaded dataset files:** 2.72 MB
|
56 |
+
- **Size of the generated dataset:** 7.63 MB
|
57 |
+
- **Total amount of disk used:** 10.35 MB
|
58 |
+
|
59 |
+
An example of 'train' looks as follows.
|
60 |
+
```
|
61 |
+
This example was too long and was cropped:
|
62 |
+
|
63 |
+
{
|
64 |
+
"abstract": "[\"Alterations of the architecture of cerebral white matter in the developing human brain can affect cortical development and res...",
|
65 |
+
"doc_id": 4983,
|
66 |
+
"structured": false,
|
67 |
+
"title": "Microstructural development of human newborn cerebral white matter assessed in vivo by diffusion tensor magnetic resonance imaging."
|
68 |
+
}
|
69 |
+
```
|
70 |
+
|
71 |
+
### Data Fields
|
72 |
+
|
73 |
+
The data fields are the same among all splits.
|
74 |
+
|
75 |
+
#### claims
|
76 |
+
|
77 |
+
- `id`: a `int32` feature.
|
78 |
+
- `claim`: a `string` feature.
|
79 |
+
- `evidence_doc_id`: a `string` feature.
|
80 |
+
- `evidence_label`: a `string` feature.
|
81 |
+
- `evidence_sentences`: a `list` of `int32` features.
|
82 |
+
- `cited_doc_ids`: a `list` of `int32` features.
|
83 |
+
|
84 |
+
#### corpus
|
85 |
+
|
86 |
+
- `doc_id`: a `int32` feature.
|
87 |
+
- `title`: a `string` feature.
|
88 |
+
- `abstract`: a `list` of `string` features.
|
89 |
+
- `structured`: a `bool` feature.
|
90 |
+
|
91 |
+
### Data Splits
|
92 |
+
|
93 |
+
#### claims
|
94 |
+
|
95 |
+
| |train|validation|test|
|
96 |
+
|------|----:|---------:|---:|
|
97 |
+
|claims| 1261| 450| 300|
|
98 |
+
|
99 |
+
#### corpus
|
100 |
+
|
101 |
+
| |train|
|
102 |
+
|------|----:|
|
103 |
+
|corpus| 5183|
|
104 |
+
|
105 |
+
## Additional Information
|
106 |
+
|
107 |
+
### Licensing Information
|
108 |
+
|
109 |
+
https://github.com/allenai/scifact/blob/master/LICENSE.md
|
110 |
+
|
111 |
+
The SciFact dataset is released under the [CC BY-NC 2.0](https://creativecommons.org/licenses/by-nc/2.0/). By using the SciFact data, you are agreeing to its usage terms.
|
112 |
+
|
113 |
+
### Citation Information
|
114 |
+
|
115 |
+
```
|
116 |
+
@inproceedings{wadden-etal-2020-fact,
|
117 |
+
title = "Fact or Fiction: Verifying Scientific Claims",
|
118 |
+
author = "Wadden, David and
|
119 |
+
Lin, Shanchuan and
|
120 |
+
Lo, Kyle and
|
121 |
+
Wang, Lucy Lu and
|
122 |
+
van Zuylen, Madeleine and
|
123 |
+
Cohan, Arman and
|
124 |
+
Hajishirzi, Hannaneh",
|
125 |
+
booktitle = "Proceedings of the 2020 Conference on Empirical Methods in Natural Language Processing (EMNLP)",
|
126 |
+
month = nov,
|
127 |
+
year = "2020",
|
128 |
+
address = "Online",
|
129 |
+
publisher = "Association for Computational Linguistics",
|
130 |
+
url = "https://aclanthology.org/2020.emnlp-main.609",
|
131 |
+
doi = "10.18653/v1/2020.emnlp-main.609",
|
132 |
+
pages = "7534--7550",
|
133 |
+
}
|
134 |
+
```
|