Update README.md

README.md

@@ -11,7 +11,7 @@ pipeline_tag: token-classification
 
 Bibtexer act like a reversed Zotero: given an unstructured list of references, the model will return a series of Bibtex entries that can be loaded in any bibliographic databases.
 
-Like all models from PleIAs Bad Data Toolbox, BibTexer has been volontary trained on diverse and challenging data sources, covering nearly all the styles featured on Zotero, as well as examples of broken text sources (line jump, digitization artifact).
+Like all models from PleIAs **Bad Data Toolbox**, BibTexer has been volontary trained on diverse and challenging data sources, covering nearly all the styles featured on Zotero, as well as examples of broken text sources (line jump, digitization artifact).
 
 BibTexer has been trained on multilingual styles and formats and should work correctly on most European languages.