Update README.md

README.md

@@ -76,7 +76,7 @@ We trained `google/mt5-small` [300 million parameters (~1.20 GB)] on the GreekSU
   * padding = ‘max_length’
   * truncation = True
 
-**Note:** T5-based models use a multi-task architecture, the prefix *‘summarize:’* was prepended in each training sample. 
+**Note:** T5-based models use a multi-task architecture, the prefix *‘summarize: ’* was prepended in each training sample. 
 
 ## Evaluation
 **Approach**|**ROUGE-1**|**ROUGE-2**|**ROUGE-L**|**BERTScore**