Classification on GAD
79.7Macro F1CLM detour
Evaluation results
| Method | Links | |
|---|---|---|
| CLM detourCtx=8192, Model Size=Large 350M, Training Tokens=50B tokens2026.05 | 79.7 | |
| PubMedBERTCtx=512, Model Size=Base2026.05 | 79.3 | |
| CLM detourCtx=8192, Model Size=Base 150M, Training Tokens=10B tokens2026.05 | 79.3 | |
| CLM detourCtx=8192, Model Size=Base 150M, Training Tokens=50B tokens2026.05 | 78.8 | |
| MLM baselineCtx=8192, Model Size=Base 150M, Training Tokens=50B tokens2026.05 | 78.7 | |
| MLM baselineCtx=8192, Model Size=Base 150M, Training Tokens=10B tokens2026.05 | 78.6 | |
| ModernBERT-baseCtx=8192, Model Size=Base2026.05 | 76.8 | |
| MLM baselineCtx=8192, Model Size=Large 350M, Training Tokens=50B tokens2026.05 | 76.4 | |
| BioClinical-ModernBERTCtx=8192, Model Size=Base2026.05 | 75.8 |