Update README.md

README.md

@@ -26,6 +26,11 @@ The CUBBITT model was published as:
 
 Popel, M., Tomkova, M., Tomek, J. et al. Transforming machine translation: a deep learning system reaches news translation quality comparable to human professionals. Nat Commun 11, 4381 (2020). https://doi.org/10.1038/s41467-020-18073-9
 
+## Annotation
+
+From the 165390 pairs, 151470 (91.58%) were considered understandable (i.e., they were not marked as "bad translation" but the translation may not be accurate enough to preserve the entailment).
+
+
 ## Inter-Annotator Agreement
 
 Two random annotators obtained the same dataset. The kappa score is 0.67 (substantial agreement). 
@@ -54,4 +59,8 @@ Full report on the agreement
 ```
 ## Dataset Formats
 
-Dataset is available as TSV and JSONL. The JSONL version only contains pairs that were *not* annotated as "bad translation".
+Dataset is available as TSV and JSONL. 
+
+The JSONL version only contains pairs that were *not* annotated as "bad translation". 
+In case of multiple annotations, only the *agreed* pairs (where both annotators agreed) are selected.
+The JSONL contains 149660 sentence pairs.