Update README.md
Browse files
README.md
CHANGED
|
@@ -20,7 +20,7 @@ BART has achieved state-of-the-art results on the CNN/Daily Mail and XSum datase
|
|
| 20 |
|
| 21 |
- On the CNN/Daily Mail dataset, BART achieved a `ROUGE-2` score of 21.28, which is the highest reported score on this dataset as of September 2021. The previous state-of-the-art model, [`google/PEGASUS`](https://huggingface.co/google/pegasus-xsum), achieved a `ROUGE-2` score of 21.00. BART also achieved state-of-the-art results on several other metrics such as `ROUGE-1` and `ROUGE-L`.
|
| 22 |
|
| 23 |
-
- On the XSum dataset, BART achieved a `ROUGE-2` score of
|
| 24 |
|
| 25 |
BART SOTA on CNN/DM
|
| 26 |
|
|
|
|
| 20 |
|
| 21 |
- On the CNN/Daily Mail dataset, BART achieved a `ROUGE-2` score of 21.28, which is the highest reported score on this dataset as of September 2021. The previous state-of-the-art model, [`google/PEGASUS`](https://huggingface.co/google/pegasus-xsum), achieved a `ROUGE-2` score of 21.00. BART also achieved state-of-the-art results on several other metrics such as `ROUGE-1` and `ROUGE-L`.
|
| 22 |
|
| 23 |
+
- On the XSum dataset, BART achieved a `ROUGE-2` score of 22.27, which is the highest reported score on this dataset as of September 2021. The previous state-of-the-art model, T5, achieved a `ROUGE-2` score of 22.06. BART also achieved state-of-the-art results on several other metrics such as `ROUGE-1` and `ROUGE-L`.
|
| 24 |
|
| 25 |
BART SOTA on CNN/DM
|
| 26 |
|