Update README.md

README.md

@@ -15,18 +15,17 @@ tags:
 - sentiment-regression
 - sentiment-classification
 - parliament
-widget:
-- text: >-
-    Poštovani potpredsjedničke Vlade i ministre hrvatskih branitelja, mislite li
-    da ste zapravo iznevjerili svoje suborce s kojima ste 555 dana prosvjedovali
-    u šatoru protiv tadašnjih dužnosnika jer ste zapravo donijeli zakon koji je
-    neprovediv, a birali ste si suradnike koji nemaju etički integritet.
+inference: false
 ---
 
 
 # Multilingual parliament sentiment regression model XLM-R-Parla-Sent
 
-This model is based on [xlm-r-parla](https://huggingface.co/classla/xlm-r-parla) and fine-tuned on manually annotated sentiment datasets from United Kingdom, Czechia, Slovakia, Slovenia, Bosnia and Herzegovina, Croatia, and Serbia.
+This model is based on [xlm-r-parla](https://huggingface.co/classla/xlm-r-parla), an XLM-R-large model additionally pre-trained on parliamentary proceedings, and fine-tuned on manually annotated sentiment datasets from Bosnia and Herzegovina, Croatia, Czechia, Serbia, Slovakia, Slovenia, and the United Kingdom.
+
+Both the additionally pre-trained model, as the training dataset are results of the [ParlaMint project](https://www.clarin.eu/parlamint). The details on the models and the dataset are described in the following publication (to be published soon):
+
+Michal Mochtak, Peter Rupnik, Nikola Ljubešić: The ParlaSent Multilingual Training Dataset for Sentiment Identification in Parliamentary Proceedings.
 
 ## Annotation schema
 
@@ -40,13 +39,11 @@ The discrete labels, present in the original dataset, were mapped to integers as
   "M_Positive": 4.0,
   "Positive": 5.0,
 ```
-Model was then fine-tuned on numeric labels and setup as regressor.
-
-
+The model was then fine-tuned on numeric labels and set up as a regressor.
 
 ## Finetuning procedure
 
-The fine-tuning procedure is described in this paper (ARXIV SUBMISSION to be added). Presumed optimal hyperparameters used are
+The fine-tuning procedure is described in the pending paper. Presumed optimal hyperparameters used are
 ```
   num_train_epochs=4,
   train_batch_size=32,
@@ -58,12 +55,12 @@ The fine-tuning procedure is described in this paper (ARXIV SUBMISSION to be add
 
 Results reported were obtained from 10 fine-tuning runs.
 
-test dataset | R^2 
+test dataset | R^2 | MAE
 --- | ---
-BCS | 0.6146 ± 0.0104
-EN | 0.6722 ± 0.0100
+BCS | 0.6146 ± 0.0104 | 
+EN | 0.6722 ± 0.0100 | 
 
-## Example
+## Usage Example
 
 With `simpletransformers==0.64.3`.
 ```python