Update README.md

README.md

@@ -142,6 +142,8 @@ subset2evaluate.evaluate.eval_subset_correlation(data_precomet[:100], data_full)
 > 0.75
 ```
 we get higher correlation.
+Note that this is not the best PreCOMET model.
+Read the paper for more information.
 
 
 This work is described in [How to Select Datapoints for Efficient Human Evaluation of NLG Models?](https://arxiv.org/abs/2501.18251).