Natural Language Inference on LiDiRus
Metric: MCC (higher is better)
Metric: MCC (higher is better)
| # | Model↕ | MCC▼ | Extra Data | Paper | Date↕ | Code |
|---|---|---|---|---|---|---|
| 1 | Human Benchmark | 0.626 | No | RussianSuperGLUE: A Russian Language Understandi... | 2020-10-29 | Code |
| 2 | ruRoberta-large finetune | 0.339 | No | - | - | - |
| 3 | ruT5-large-finetune | 0.32 | No | - | - | - |
| 4 | ruT5-base-finetune | 0.267 | No | - | - | - |
| 5 | ruBert-large finetune | 0.235 | No | - | - | - |
| 6 | RuGPT3Large | 0.231 | No | - | - | - |
| 7 | ruBert-base finetune | 0.224 | No | - | - | - |
| 8 | SBERT_Large_mt_ru_finetuning | 0.218 | No | - | - | - |
| 9 | SBERT_Large | 0.209 | No | - | - | - |
| 10 | RuBERT plain | 0.191 | No | - | - | - |
| 11 | Multilingual Bert | 0.189 | No | - | - | - |
| 12 | RuBERT conversational | 0.178 | No | - | - | - |
| 13 | heuristic majority | 0.147 | No | Unreasonable Effectiveness of Rule-Based Heurist... | 2021-05-03 | - |
| 14 | YaLM 1.0B few-shot | 0.124 | No | - | - | - |
| 15 | RuGPT3XL few-shot | 0.096 | No | - | - | - |
| 16 | MT5 Large | 0.061 | No | mT5: A massively multilingual pre-trained text-t... | 2020-10-22 | Code |
| 17 | Baseline TF-IDF1.1 | 0.06 | No | RussianSuperGLUE: A Russian Language Understandi... | 2020-10-29 | Code |
| 18 | RuGPT3Medium | 0.01 | No | - | - | - |
| 19 | Golden Transformer | 0 | No | - | - | - |
| 20 | Random weighted | 0 | No | - | - | - |
| 21 | majority_class | 0 | No | - | - | - |
| 22 | RuGPT3Small | -0.013 | No | - | - | - |