Update README.md

README.md

@@ -62,8 +62,8 @@ The model may reflect biases present in the training data, such as jurisdictiona
 
 ### Recommendations
 
-- Outputs should always be reviewed by a legal expert.
-- Avoid using for legal tasks where complete precision is mandatory.
+- A legal expert should always review outputs.
+- Avoid using it for legal tasks where complete precision is mandatory.
 
 
 ### Training Data
@@ -99,9 +99,8 @@ The model may reflect biases present in the training data, such as jurisdictiona
 - Validation was performed on the `validation` split of the Multi-LexSum dataset, consisting of 4,818 examples.
 
 #### Metrics
-- **ROUGE-1:** 0.49
-- **ROUGE-2:** 0.35
-- **ROUGE-L:** 0.49
+- **bert_score Short Summary Precision :** 0.84
+- **bert_score Long Summary Precision :** 0.81
 
 ### Results
 - The model produces reliable short and long summaries for legal documents, maintaining coherence and relevance.