Update README.md

README.md

@@ -128,7 +128,7 @@ F1 score: A weighted average of precision and recall, which measures the overlap
 | RoBERTa-Base | 124,056,578 | 8:50:29 | 18:59 | 2:11 | 69.06 | 80.08 | 55.53 | 66.49 |
 | RoBERTa-Large | 354,312,194 | 29:16:06 | 1:01:10 | 7:04 | 74.08 | 84.38 | 62.20 | 72.88 |
 
-\* Training times aren't perfectly comparable as TinyRoBERTa was distilled from [VMware/roberta-large-mrqa](https://huggingface.co/VMware/roberta-large-mrqa) that was already trained on MRQA
+\* TinyRoBERTa's training time isn't directly comparable to the other models since it was distilled from [VMware/roberta-large-mrqa](https://huggingface.co/VMware/roberta-large-mrqa) that was already trained on MRQA.
 
 # Limitations and Bias