Update README.md

README.md

@@ -129,12 +129,6 @@ The fine-tuning script can be accessed [here](https://github.com/jgrosjean-mathe
 
 <!-- This section describes the evaluation protocols and provides the results. -->
 
-#### Baseline
-
-The first baseline is [distiluse-base-multilingual-cased](https://www.sbert.net/examples/training/multilingual/README.html), a high-performing Sentence Transformer model that is able to process German, French and Italian (and more).
-
-The second baseline uses mean pooling embeddings from the last hidden state of the original swissbert model.
-
 #### Testing Data
 
 <!-- This should link to a Dataset Card if possible. -->
@@ -170,4 +164,10 @@ Making use of an unsupervised training approach, Swissbert for Sentence Embeddin
 | Semantic Similarity IT | 83.00    | - `     | 84.00          |    -    |**89.80**    | -       |
 | Text Classification DE | 95.76    |**91.99**| 94.70          |  89.43  |  95.61      | 91.20   |
 | Text Classification FR | 94.55    | 88.52   | 95.30          |**89.91**|  94.55      | 89.82   |
-| Text Classification IT | 93.48    | 88.29   | 94.85          |  90.36  |  95.91      |**92.05**|
+| Text Classification IT | 93.48    | 88.29   | 94.85          |  90.36  |  95.91      |**92.05**|
+
+#### Baselines
+
+The first baseline uses mean pooling embeddings from the last hidden state of the original swissbert model.
+
+The second baseline is [distiluse-base-multilingual-cased](https://www.sbert.net/examples/training/multilingual/README.html), a high-performing Sentence Transformer model that is able to process German, French and Italian (and more).