Update README.md
Browse files
README.md
CHANGED
|
@@ -1,3 +1,18 @@
|
|
| 1 |
---
|
| 2 |
license: unknown
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 3 |
---
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
---
|
| 2 |
license: unknown
|
| 3 |
+
task_categories:
|
| 4 |
+
- text-classification
|
| 5 |
+
language:
|
| 6 |
+
- en
|
| 7 |
+
size_categories:
|
| 8 |
+
- 10K<n<100K
|
| 9 |
---
|
| 10 |
+
A dataset of contradictory sentence pairs.
|
| 11 |
+
|
| 12 |
+
This is the dataset from a Master's thesis: https://repositorio-aberto.up.pt/bitstream/10216/129029/2/415679.pdf. The name of the dataset comes from the title of the thesis. You can find some documentation for the dataset in the following repository: https://github.com/BeatrizBaldaia/sentence-pair-contradictions/tree/master
|
| 13 |
+
|
| 14 |
+
Originally collated by Beatriz Souto de Sá Baldaia, I take no credit for that aspect, I just processed and uploaded it here because I needed to train a sentence contradiction model.
|
| 15 |
+
|
| 16 |
+
This dataset is a merge of several others, so I'm not sure what the license should be.
|
| 17 |
+
|
| 18 |
+
I think there is quite a lot of political content in the data, but models trained from it do seem to generalise
|