| Benchmark | Model | Rank | Results |
|---|---|---|---|
| open-domain-question-answering-on-eli5 | BART | #3 | Rouge-L: 24.3Rouge-1: 30.6Rouge-2: 6.2 |
| text-summarization-on-x-sum | BART | #6 | ROUGE-1: 45.14ROUGE-2: 22.27ROUGE-3: 37.25 |
| question-answering-on-squad20-dev | BART | #7 | F1: 89.2 |
| natural-language-inference-on-qnli | BART | #12 | Accuracy: 94.9% |
| abstractive-text-summarization-on-cnn-daily-mail | BART | #18 | ROUGE-1: 44.16ROUGE-2: 21.28ROUGE-L: 40.90 |
| question-answering-on-squad11-dev | BART Base (with text infilling) | #42 | F1: 90.8 |