📊 Showing 6 results | 📏 Metric: Accuracy
| Rank | Model | Paper | Accuracy | Date | Code |
|---|---|---|---|---|---|
| 1 | PaLM 2 (few-shot) | PaLM 2 Technical Report | 94.40 | 2023-05-17 | 📦 eternityyw/tram-benchmark |
| 2 | mT0-13B | Crosslingual Generalization through Multitask Finetuning | 84.45 | 2022-11-03 | 📦 bigscience-workshop/xmtf |
| 3 | RoBERTa Large (translate test) 📚 | XCOPA: A Multilingual Dataset for Causal Commonsense Reasoning | 76.05 | 2020-05-01 | 📦 cambridgeltl/xcopa |
| 4 | BLOOMZ | Crosslingual Generalization through Multitask Finetuning | 75.50 | 2022-11-03 | 📦 bigscience-workshop/xmtf |
| 5 | MAD-X Base 📚 | MAD-X: An Adapter-Based Framework for Multi-Task Cross-Lingual Transfer | 60.94 | 2020-04-30 | 📦 Adapter-Hub/adapter-transformers 📦 cambridgeltl/xcopa 📦 aaronsom/wmt21-qe-tudarmstadt |
| 6 | mGPT | mGPT: Few-Shot Learners Go Multilingual | 55.50 | 2022-04-15 | 📦 ai-forever/mgpt |