📊 Showing 6 results | 📏 Metric: Accuracy
| Rank | Model | Paper | Accuracy | Date | Code |
|---|---|---|---|---|---|
| 1 | Human Benchmark | RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark | 0.92 | 2020-10-29 | 📦 RussianNLP/RussianSuperGLUE 📦 RussianNLP/MOROCCO |
| 2 | MT5 Large | mT5: A massively multilingual pre-trained text-to-text transformer | 0.66 | 2020-10-22 | 📦 huggingface/transformers 📦 google-research/multilingual-t5 📦 google-research/byt5 |
| 3 | heuristic majority | Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks | 0.64 | 2021-05-03 | - |
| 4 | Baseline TF-IDF1.1 | RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark | 0.62 | 2020-10-29 | 📦 RussianNLP/RussianSuperGLUE 📦 RussianNLP/MOROCCO |
| 5 | Random weighted | Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks | 0.52 | 2021-05-03 | - |
| 6 | majority_class | Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks | 0.50 | 2021-05-03 | - |