Classification on CLIP 200 random samples (test)
0.24Macro F1 ScoreQwen3-30B
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Qwen3-30BModel Category=New-gen general (Mnew)2026.01 | 0.24 | 0.54 | |
| CAPTComponents=Mnew + Mold-clin + Mold2026.01 | 0.224 | 0.58 | |
| UniTEComponents=Mnew + Mold-clin2026.01 | 0.217 | 0.56 | |
| MeLLaMA-13B-chatModel Category=Old-gen clinical (Mold-clin)2026.01 | 0.192 | 0.785 | |
| Proxy TuningComponents=Mold-L + Mold-clin + Mold2026.01 | 0.143 | 0.42 | |
| LLaMA-2-70B-chatModel Category=Old-gen large general (Mold-L)2026.01 | 0.097 | 0.205 |