Table 3.
Category-specific accuracy results.
| Subject | Calculation | Diagnosis | Ethics | Management | Pharmacology | Physiology |
|---|---|---|---|---|---|---|
| Fellow 1 | 75.00 | 70.59 | 100.00 | 76.67 | 88.89 | 65.52 |
| Fellow 2 | 50.00 | 88.23 | 66.67 | 86.67 | 77.78 | 79.31 |
| Phi-4 | 25.00 | 76.47 | 100.00 | 73.33 | 100 | 65.51 |
| Llama-3.3-70B | 50.00 | 76.47 | 100.00 | 70.00 | 88.89 | 72.41 |
Category-specific accuracy comparison for two pediatric critical care fellows and the two best-performing LLMs.