Medical Image Classification on MediConfusion
0.4375Set AccuracyCCI
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| CCIBase Model=GEMINI 2.0 PRO2026.02 | 0.4375 | 0.642 | 0.3693 | 3.69 | |
| CCIBase Model=GPT-o12026.02 | 0.3295 | 0.5994 | 0.4261 | 4.26 | |
| DIRECT (SINGLE-SHOT)Base Model=GEMINI 2.0 PRO2026.02 | 0.2955 | 0.6193 | 0.6705 | — | |
| CCIBase Model=GPT-4o2026.02 | 0.2784 | 0.5511 | 0.4943 | 4.83 | |
| RANDOM GUESSING2026.02 | 0.25 | 0.5 | 0.5 | — | |
| DIRECT (SINGLE-SHOT)Base Model=GPT-o12026.02 | 0.2169 | 0.5795 | 0.7299 | — | |
| DIRECT (SINGLE-SHOT)Base Model=GPT-4o2026.02 | 0.1875 | 0.5625 | 0.75 | — |