Classification on AI-READI Scenario C — Dual-Class Undersampling (train)
21.47ACCLLaMA*
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| LLaMA*Augmentation=CF Added, Prompting=Fine-tuned2026.01 | 21.47 | 19.67 | 22.03 | 20 | 26.87 | |
| BioMistral*Augmentation=CF Added, Prompting=Fine-tuned2026.01 | 15.99 | 16.39 | 16.95 | 16.67 | 22.39 | |
| GPT-4Augmentation=CF Added, Prompting=Zero-shot2026.01 | 6.27 | 4.92 | 5.08 | 5 | 8.96 | |
| LLaMAAugmentation=CF Added, Prompting=Zero-shot2026.01 | 5.9 | 4.92 | 6.78 | 5 | 7.46 | |
| BioMistralAugmentation=CF Added, Prompting=Zero-shot2026.01 | 4.55 | 1.64 | 3.39 | 1.67 | 4.48 | |
| NNData Condition=Model trained on reduced data2026.01 | -11.14 | -10.29 | -13.24 | -11.76 | -14.1 |