| Benchmark | Model | Rank | Results |
|---|---|---|---|
| cross-lingual-natural-language-inference-on-xnli | ByT5 XXL | #1 | Accuracy: 83.7 |
| cross-lingual-natural-language-inference-on-xnli | ByT5 Small | #4 | Accuracy: 69.1 |
| cross-lingual-question-answering-on-tydiqa-goldp | ByT5 (fine-tuned) | #1 | EM: 81.9 |
| cross-lingual-question-answering-on-tydiqa-goldp | ByT5 XXL | #4 | EM: 60.0F1: 75.3 |
| cross-lingual-question-answering-on-xquad | ByT5 XXL | #1 | EM: 63.6F1: 79.7 |
| extreme-summarization-on-gem-xsum | ByT5 | #4 | BLEU score: 15.3 |
| extreme-summarization-on-gem-xsum | mT5 | #5 | BLEU score: 14.3 |