Datasets:
Tasks:
Text Classification
Modalities:
Text
Formats:
json
Languages:
English
Size:
10K - 100K
ArXiv:
License:
Update readme.md
Browse files
readme.md
CHANGED
|
@@ -0,0 +1,8 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
This is the data used in the paper [Large Language Model as Attributed Training Data Generator: A Tale of Diversity and Bias](https://github.com/yueyu1030/AttrPrompt).
|
| 2 |
+
|
| 3 |
+
- `label.txt`: the label name for each class
|
| 4 |
+
- `train.jsonl`: The original training set.
|
| 5 |
+
- `valid.jsonl`: The original validation set.
|
| 6 |
+
- `test.jsonl`: The original test set.
|
| 7 |
+
- `simprompt.jsonl`: The training data generated by the simple prompt.
|
| 8 |
+
- `attrprompt.jsonl`: The training data generated by the attributed prompt.
|