Update README.md
Browse files
README.md
CHANGED
|
@@ -33,6 +33,7 @@ Karibu is a DeBERTa-based classifier that automatically assigns CEFR language pr
|
|
| 33 |
## Training Details
|
| 34 |
|
| 35 |
- Training Data: 9,000 synthetic samples
|
|
|
|
| 36 |
|
| 37 |
- Source: French press articles + Wikimedia content
|
| 38 |
- Processing: Sequential text simplification using an open source model (to be published)
|
|
@@ -46,9 +47,6 @@ Karibu is a DeBERTa-based classifier that automatically assigns CEFR language pr
|
|
| 46 |
🤖 [Explore the Celadon model](https://huggingface.co/PleIAs/celadon)
|
| 47 |
|
| 48 |
|
| 49 |
-
🔍 [Explore the full dataset](https://huggingface.co/datasets/PleIAs/KaribuAI/viewer/default)
|
| 50 |
-
|
| 51 |
-
|
| 52 |
## levels
|
| 53 |
- Manual verification using CEFR framework criteria
|
| 54 |
- Statistical validation using Louvain word-level classification
|
|
|
|
| 33 |
## Training Details
|
| 34 |
|
| 35 |
- Training Data: 9,000 synthetic samples
|
| 36 |
+
🔍 [Explore the full dataset](https://huggingface.co/datasets/PleIAs/KaribuAI/viewer/default)
|
| 37 |
|
| 38 |
- Source: French press articles + Wikimedia content
|
| 39 |
- Processing: Sequential text simplification using an open source model (to be published)
|
|
|
|
| 47 |
🤖 [Explore the Celadon model](https://huggingface.co/PleIAs/celadon)
|
| 48 |
|
| 49 |
|
|
|
|
|
|
|
|
|
|
| 50 |
## levels
|
| 51 |
- Manual verification using CEFR framework criteria
|
| 52 |
- Statistical validation using Louvain word-level classification
|