Datasets:
Tasks:
Question Answering
Modalities:
Text
Formats:
json
Sub-tasks:
extractive-qa
Languages:
Catalan
Size:
1K - 10K
ArXiv:
License:
Commit ·
26fff6d
1
Parent(s): 5f78245
Update README.md
Browse files
README.md
CHANGED
|
@@ -95,8 +95,8 @@ The dataset is in Catalan (`ca-CA`).
|
|
| 95 |
Follows [Rajpurkar, Pranav et al., 2016](http://arxiv.org/abs/1606.05250) for squad v1 datasets.
|
| 96 |
|
| 97 |
- `id` (str): Unique ID assigned to the question.
|
| 98 |
-
- `title` (str): Title of the
|
| 99 |
-
- `context` (str):
|
| 100 |
- `question` (str): Question.
|
| 101 |
- `answers` (list): List of answers to the question, each containing:
|
| 102 |
- `text` (str): Span text answering to the question.
|
|
|
|
| 95 |
Follows [Rajpurkar, Pranav et al., 2016](http://arxiv.org/abs/1606.05250) for squad v1 datasets.
|
| 96 |
|
| 97 |
- `id` (str): Unique ID assigned to the question.
|
| 98 |
+
- `title` (str): Title of the VilaWeb article.
|
| 99 |
+
- `context` (str): VilaWeb section text.
|
| 100 |
- `question` (str): Question.
|
| 101 |
- `answers` (list): List of answers to the question, each containing:
|
| 102 |
- `text` (str): Span text answering to the question.
|