Diagnostic report generation on PPGL
3.6Diagnostic AccuracyPPGL-Swarm
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| PPGL-SwarmInput=Slide+Gene+Table2026.03 | 3.6 | 3 | 2.7 | 3.5 | 3.2 | |
| TITANInput=Slide+Gene+Table2026.03 | 3.4 | 2.6 | 2.4 | 3.1 | 2.88 | |
| SlideChatInput=Slide+Gene+Table2026.03 | 3.1 | 2.8 | 2.6 | 3.3 | 2.95 | |
| MedDrInput=Patch+Gene+Table2026.03 | 2.9 | 2.5 | 2.3 | 3.2 | 2.73 | |
| GPT-4oInput=Patch+Gene+Table2026.03 | 2.8 | 2.3 | 2.1 | 3.1 | 2.58 | |
| Claude-4.5-SonnetInput=Patch+Gene+Table2026.03 | 2.7 | 2.2 | 2 | 3.2 | 2.53 | |
| GPT-4oInput=Thumbnail+Gene+Table2026.03 | 2.2 | 2 | 1.9 | 3 | 2.28 | |
| LLaVA-MedInput=Patch+Gene+Table2026.03 | 2.1 | 1.8 | 1.7 | 2.6 | 2.05 |