Datasets:
Tasks:
Text Generation
Modalities:
Text
Formats:
json
Languages:
Ukrainian
Size:
10K - 100K
License:
Update README.md
Browse files
README.md
CHANGED
@@ -1,8 +1,49 @@
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
|
|
|
|
|
|
|
|
3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
4 |
## Dataset Summary
|
5 |
-
This is the dataset
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
6 |
|
7 |
# Dataset Structure
|
8 |
The dataset is in JSON format.
|
|
|
1 |
---
|
2 |
license: cc-by-nc-4.0
|
3 |
+
task_categories:
|
4 |
+
- text-generation
|
5 |
+
language:
|
6 |
+
- uk
|
7 |
---
|
8 |
+
|
9 |
+
# Dataset Card for Spivavtor
|
10 |
+
|
11 |
+
## Paper: Spivavtor: An Instruction Tuned Ukrainian Text Editing Model
|
12 |
+
|
13 |
+
## Authors: Aman Saini, Artem Chernodub, Vipul Raheja, Vivek Kulkarni
|
14 |
+
|
15 |
## Dataset Summary
|
16 |
+
This is the dataset used to train all Spivavtor models. It contains data for 4 tasks - Grammatical Error Correction (GEC), Simplification, Coherence and Paraphrasing.
|
17 |
+
The specific details are as follows:
|
18 |
+
<table>
|
19 |
+
<tr>
|
20 |
+
<th>Task</th>
|
21 |
+
<th>Examples in Training data</th>
|
22 |
+
<th>Examples in Validation data</th>
|
23 |
+
</tr>
|
24 |
+
<tr>
|
25 |
+
<td>GEC</td>
|
26 |
+
<td>27,929</td>
|
27 |
+
<td>3,103</td>
|
28 |
+
</tr>
|
29 |
+
<tr>
|
30 |
+
<td>Simplification</td>
|
31 |
+
<td>11,501</td>
|
32 |
+
<td>1,278</td>
|
33 |
+
</tr>
|
34 |
+
<tr>
|
35 |
+
<td>Coherence</td>
|
36 |
+
<td>9,278</td>
|
37 |
+
<td>1,031</td>
|
38 |
+
</tr>
|
39 |
+
<tr>
|
40 |
+
<td>Paraphrasing</td>
|
41 |
+
<td>14,076</td>
|
42 |
+
<td>1,564</td>
|
43 |
+
</tr>
|
44 |
+
</table>
|
45 |
+
|
46 |
+
Full details of the dataset can be found in our paper.
|
47 |
|
48 |
# Dataset Structure
|
49 |
The dataset is in JSON format.
|