| Paper | Code | Accuracy | ModelName | ReleaseDate |
|---|---|---|---|---|
| RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark | ✓ Link | 0.92 | Human Benchmark | 2020-10-29 |
| []() | 0.871 | Golden Transformer | ||
| []() | 0.801 | ruRoberta-large finetune | ||
| []() | 0.747 | ruT5-large-finetune | ||
| []() | 0.704 | ruBert-large finetune | ||
| []() | 0.703 | ruBert-base finetune | ||
| []() | 0.692 | ruT5-base-finetune | ||
| []() | 0.654 | RuGPT3Large | ||
| []() | 0.642 | RuBERT plain | ||
| []() | 0.64 | RuBERT conversational | ||
| []() | 0.637 | SBERT_Large_mt_ru_finetuning | ||
| []() | 0.637 | SBERT_Large | ||
| []() | 0.617 | Multilingual Bert | ||
| []() | 0.605 | YaLM 1.0B few-shot | ||
| []() | 0.573 | RuGPT3XL few-shot | ||
| mT5: A massively multilingual pre-trained text-to-text transformer | ✓ Link | 0.561 | MT5 Large | 2020-10-22 |
| Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks | 0.549 | heuristic majority | 2021-05-03 | |
| Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks | 0.513 | majority_class | 2021-05-03 | |
| []() | 0.505 | RuGPT3Medium | ||
| []() | 0.488 | RuGPT3Small | ||
| Unreasonable Effectiveness of Rule-Based Heuristics in Solving Russian SuperGLUE Tasks | 0.483 | Random weighted | 2021-05-03 | |
| RussianSuperGLUE: A Russian Language Understanding Evaluation Benchmark | ✓ Link | 0.471 | Baseline TF-IDF1.1 | 2020-10-29 |