Update README.md
Browse files
README.md
CHANGED
|
@@ -249,6 +249,8 @@ While the parent dataset (`ando55/WikiSQE`) contains **every** sentence flagged
|
|
| 249 |
* **27 dataset *groups*** (20 frequent quality labels + 5 Quality type categories + 2 Coarse groups)
|
| 250 |
* **3 standard splits per group** (`train`, `val`, `test`) – for example `citation/train`, `citation/val`, …
|
| 251 |
|
|
|
|
|
|
|
| 252 |
> **Need the full dump?** Head to [https://huggingface.co/datasets/ando55/WikiSQE](https://huggingface.co/datasets/ando55/WikiSQE).
|
| 253 |
|
| 254 |
---
|
|
|
|
| 249 |
* **27 dataset *groups*** (20 frequent quality labels + 5 Quality type categories + 2 Coarse groups)
|
| 250 |
* **3 standard splits per group** (`train`, `val`, `test`) – for example `citation/train`, `citation/val`, …
|
| 251 |
|
| 252 |
+
Each split blends **labeled** and **unlabeled** sentences at a **1 : 1 ratio** to support semi-supervised and positive/negative training paradigms.
|
| 253 |
+
|
| 254 |
> **Need the full dump?** Head to [https://huggingface.co/datasets/ando55/WikiSQE](https://huggingface.co/datasets/ando55/WikiSQE).
|
| 255 |
|
| 256 |
---
|