Update README.md

README.md

@@ -12,7 +12,7 @@ datasets:
 
 ## BART Legal Spanish ⚖️
 
-**BART Legal** (base) is an BART-like model trained on [A collection of corpora of Spanish legal domain](https://zenodo.org/record/5495529#.YZItp3vMLJw).
+**BART Legal Spanish** (base) is a BART-like model trained on [A collection of corpora of Spanish legal domain](https://zenodo.org/record/5495529#.YZItp3vMLJw).
 
 BART is a transformer encoder-decoder (seq2seq) model with a bidirectional (BERT-like) encoder and an autoregressive (GPT-like) decoder. BART is pre-trained by (1) corrupting text with an arbitrary noising function and (2) learning a model to reconstruct the original text.