Regenerate model card

README.md

@@ -27,12 +27,9 @@ The Basque model uses ``FacebookAI/xlm-roberta-base`` as its base encoder and is
 
 ## Evaluation
 
-| Metric | F1 |
-| ------ | -- |
-| Tokens | 99.99 |
-| Words | 99.98 |
-| Sentences | 100.00 |
-
+| Metric | Tokens | Words | Sentences |
+| ------ | ------ | ----- | --------- |
+| F1 | 99.99 | 99.98 | 100.00 |
 
 
 ## Usage
@@ -50,7 +47,7 @@ from combo_seg import ComboSeg
 nlp = ComboSeg("Basque")
 
 # Segment raw text — returns Document with hierarchy: Document -> Turn -> Sentence -> Token
-doc = nlp("Example sentence.")
+doc = nlp("Azeri marroi bizkorrak txakur alferraren gainetik salto egiten du.")
 
 # Inspect results
 for turn in doc.turns:
@@ -69,7 +66,7 @@ Or load directly from HuggingFace:
 from combo_seg import ComboSeg
 
 nlp = ComboSeg.from_pretrained("clarin-pl/combo-seg-xlm-roberta-base-basque-bdt-ud2.17")
-doc = nlp("Example sentence.")
+doc = nlp("Azeri marroi bizkorrak txakur alferraren gainetik salto egiten du.")
 ```
 
 ## License
@@ -83,7 +80,7 @@ The training data license: cc-by-nc-sa-3.0 is derived from the Universal Depende
 
 If you use this model, please cite:
 
-Ulewicz, M., & Wróblewska, A. (2026). *COMBO-SEG Models Trained on UD v2.17*. Zenodo. https://doi.org/10.5281/zenodo.19651441
+Ulewicz, M., & Wróblewska, A. (2026). *COMBO-SEG Models Trained on UD v2.17*. https://doi.org/10.5281/zenodo.19651441
 
 ```bibtex
 @software{combo_seg_2026,
@@ -96,8 +93,6 @@ Ulewicz, M., & Wróblewska, A. (2026). *COMBO-SEG Models Trained on UD v2.17*. Z
 }
 ```
 
-
-
 ## Resources
 
 - COMBO-SEG: [https://gitlab.clarin-pl.eu/syntactic-tools/combo-seg](https://gitlab.clarin-pl.eu/syntactic-tools/combo-seg)