Emotional valence classification on OB-GYN (test)
96.62RecallGatorTron
Evaluation results
| Method | Links | ||
|---|---|---|---|
| GatorTronInput=LP, Training Strategy=Prompt FT2026.02 | 96.62 | — | |
| GatorTronInput=Anchored, Training Strategy=Prompt FT2026.02 | 96.37 | — | |
| Llama-3.2-1BInput=LP, Training Strategy=Instruction FT2026.02 | 96.35 | — | |
| GatorTronInput=Anchored, Training Strategy=Standard FT2026.02 | 95.76 | — | |
| Llama-3.2-1BInput=ICL + LP, Training Strategy=Instruction FT2026.02 | 95.1 | — | |
| BERTInput=LP, Training Strategy=Prompt FT2026.02 | 95.06 | — | |
| BERTInput=Anchored, Training Strategy=Prompt FT2026.02 | 95.04 | — | |
| Med42-8BInput=ICL + LP, Training Strategy=Instruction FT2026.02 | 93.52 | — | |
| Med42-8BInput=LP, Training Strategy=Instruction FT2026.02 | 92.32 | — | |
| BERTInput=Anchored, Training Strategy=Standard FT2026.02 | 91.32 | — | |
| Med42-70BInput=LP, Training Strategy=Optimized ICL2026.02 | 90.5 | — | |
| Llama-3.3-70BInput=LP, Training Strategy=Optimized ICL2026.02 | 85.32 | — | |
| Med42-70BInput=LP, Training Strategy=Zero-shot2026.02 | 67.84 | — | |
| Llama-3.3-70BInput=LP, Training Strategy=Zero-shot2026.02 | 66.6 | — | |
| Med42-70BInput=Anchored, Training Strategy=Zero-shot2026.02 | 42.8 | — | |
| Llama-3.3-70BInput=Anchored, Training Strategy=Zero-shot2026.02 | 37 | — | |
| BERTInput=Anchored, Training Strategy=Standard FT2026.02 | — | 91.93 | |
| BERTInput=Anchored, Training Strategy=Prompt FT2026.02 | — | 91.72 | |
| BERTInput=LP, Training Strategy=Prompt FT2026.02 | — | 94.32 | |
| GatorTronInput=Anchored, Training Strategy=Standard FT2026.02 | — | 93.61 | |
| GatorTronInput=Anchored, Training Strategy=Prompt FT2026.02 | — | 93.88 | |
| GatorTronInput=LP, Training Strategy=Prompt FT2026.02 | — | 95.03 | |
| Llama-3.2-1BInput=ICL + LP, Training Strategy=Instruction FT2026.02 | — | 94.51 | |
| Llama-3.2-1BInput=LP, Training Strategy=Instruction FT2026.02 | — | 94.44 | |
| Llama-3.3-70BInput=Anchored, Training Strategy=Zero-shot2026.02 | — | 42.2 | |
| Llama-3.3-70BInput=LP, Training Strategy=Zero-shot2026.02 | — | 76.11 | |
| Llama-3.3-70BInput=LP, Training Strategy=Optimized ICL2026.02 | — | 83.76 | |
| Med42-70BInput=Anchored, Training Strategy=Zero-shot2026.02 | — | 61.89 | |
| Med42-70BInput=LP, Training Strategy=Zero-shot2026.02 | — | 71.47 | |
| Med42-70BInput=LP, Training Strategy=Optimized ICL2026.02 | — | 85.1 | |
| Med42-8BInput=ICL + LP, Training Strategy=Instruction FT2026.02 | — | 94.69 | |
| Med42-8BInput=LP, Training Strategy=Instruction FT2026.02 | — | 90.99 |