Update README.md

README.md

@@ -148,13 +148,11 @@ learning rate warmup for 10,000 steps and linear decay of the learning rate afte
 
 ## Evaluation results
 
-When fine-tuned on downstream tasks, this model achieves the following results:
-
-Glue test results:
+According to [the official paper](https://arxiv.org/abs/2105.03824) (*cf.* with Table 1 on page 7), this model achieves the following performance on the GLUE test data:
 
 | Task | MNLI-(m/mm) | QQP  | QNLI | SST-2 | CoLA | STS-B | MRPC | RTE  | Average |
 |:----:|:-----------:|:----:|:----:|:-----:|:----:|:-----:|:----:|:----:|:-------:|
-|      | 72/73   | 84 | 80 | 95  | 69 | 79  | 76 | 63| 76.7    |
+|      | 72/73   | 83 | 80 | 95  | 69 | 79  | 76 | 63| 76.7    |
 
 
 The following table contains test results on the HuggingFace model in comparison with [bert-base-cased](https://hf.co/models/bert-base-cased). The training was done on a single 16GB NVIDIA Tesla V100 GPU. For MRPC/WNLI, the models were trained for 5 epochs, while for other tasks, the models were trained for 3 epochs. Please refer to the checkpoints linked with the scores. The sequence length used for 512 with batch size 16 and learning rate 2e-5.