Hindi
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Classification and Explainability | Hindi | Accuracy87.45 | 14 | |
| Tokenization | Hindi hi | NSL Score2.65 | 10 | |
| Language Steering | Hindi (test) | Delta Mass0.85 | 7 | |
| Formality Transfer | Hindi | ACC70 | 7 | |
| Formality Transfer Magnitude Control | Hindi | CALIB69.6 | 7 | |
| Question Answering | Hindi (HI) | Accuracy (ACC)91 | 6 | |
| Tokenization Efficiency | Hindi hi | Bytes per Token7.37 | 6 | |
| Token Fertility | Hindi | Fertility2.67 | 4 | |
| Named Entity Recognition | Hindi (test) | F1 Score65.8 | 4 | |
| Code-mixing addition | Hindi | ACC62.5 | 3 | |
| Disfluency Correction | Hindi Real Data | Proposed Rate28 | 2 | |
| Disfluency Correction | Hindi Manually Edited | Proposed (%)18.1 | 2 | |
| Accent Shifting | Hindi (evaluation set) | Target Accent Probability24.2 | 2 |
Showing 13 of 13 rows