Language Identification on UDHR Full (366 labels) (test)
87.1F1 ScoreGlotLID-M
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GlotLID-MArchitecture=fastText, Training Corpus=GlotLID-C2026.02 | 87.1 | 0.0137 | |
| UniLIDTokenizer=UnigramLM, Training Corpus=GlotLID-C2026.02 | 85.9 | 0.0143 | |
| fastTextTraining Corpus=GlotLID-C2026.02 | 85.5 | 0.0187 | |
| UniLID-Mistral-NemoTokenizer=Mistral-Nemo-Base-2407, Training Corpus=GlotLID-C2026.02 | 83.6 | 0.0179 |