zouharvi commited on
Commit
2239f67
·
verified ·
1 Parent(s): fdc1f59

Update README.md

Browse files
Files changed (1) hide show
  1. README.md +1 -0
README.md CHANGED
@@ -142,6 +142,7 @@ subset2evaluate.evaluate.eval_subset_correlation(data_precomet[:100], data_full)
142
  > 0.81
143
  ```
144
  we get higher correlation.
 
145
 
146
 
147
  This work is described in [How to Select Datapoints for Efficient Human Evaluation of NLG Models?](https://arxiv.org/abs/2501.18251).
 
142
  > 0.81
143
  ```
144
  we get higher correlation.
145
+ You can expect a bigger effect on a larger scale, as described in the paper.
146
 
147
 
148
  This work is described in [How to Select Datapoints for Efficient Human Evaluation of NLG Models?](https://arxiv.org/abs/2501.18251).