Disease Diagnosis on Open-I
99.22AccuracyOracle
Evaluation results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| OracleCategory=Tool Selection Method2026.02 | 99.22 | 58 | 94.8 | 61.11 | 60.71 | — | |
| MedMIX2026.05 | 94.63 | 72.46 | — | — | 95.7 | 77.8 | |
| Note-LargeModality=Note, Model Type=Teacher Large2026.05 | 93.63 | 69.8 | — | — | 93.85 | 74.81 | |
| M3Care2026.05 | 92.65 | 66.12 | — | — | 94.19 | 72.7 | |
| AdaCoMed2026.05 | 91.93 | 62.97 | — | — | 92.6 | 67.72 | |
| ToolSelectCategory=Tool Selection Method2026.02 | 87.64 | 43.8 | 48.55 | 57.31 | 45.09 | — | |
| MeanAvgFusion=Mean Average2026.05 | 87.64 | 72.01 | — | — | 89.04 | 54.85 | |
| OneLLM2026.05 | 87.32 | 71.24 | — | — | 87.84 | 56.35 | |
| ConcatFusion=Concatenation2026.05 | 85.57 | 68.29 | — | — | 86.22 | 48.17 | |
| AttentionFusion=Attention-based2026.05 | 84.49 | 65.63 | — | — | 84.77 | 46.27 | |
| R-AllCategory=Tool Candidate Model2026.02 | 84.08 | 6.61 | 47.12 | 14.49 | 40.84 | — | |
| MaxFusion=Max Pooling2026.05 | 83.61 | 66.21 | — | — | 84.77 | 45.66 | |
| RouterDCCategory=Tool Selection Method2026.02 | 81.52 | 8.51 | 33.24 | 19.47 | 38.11 | — | |
| E-T-C14Category=Tool Candidate Model2026.02 | 78.21 | 7.37 | 31.84 | 22.9 | 37.19 | — | |
| R-MKPCategory=Tool Candidate Model2026.02 | 78.17 | 8.21 | 20.73 | 23.83 | 37.73 | — | |
| R-BioViLCategory=Tool Candidate Model2026.02 | 78.13 | 6.82 | 30.45 | 22.55 | 37.69 | — | |
| E-B-C14Category=Tool Candidate Model2026.02 | 78.13 | 7.97 | 21.2 | 23.35 | 37.76 | — | |
| D-MV2Category=Tool Candidate Model2026.02 | 78.1 | 9.61 | 37.9 | 25.46 | 38.27 | — | |
| E-S-C14Category=Tool Candidate Model2026.02 | 78.03 | 8.01 | 31.92 | 23.46 | 36.55 | — | |
| R-MGCACategory=Tool Candidate Model2026.02 | 77.96 | 8.61 | 27.39 | 24.29 | 37.46 | — | |
| Front-LargeModality=Front CXR, Model Type=Teacher Large2026.05 | 76.26 | 34.27 | — | — | 74.82 | 31.55 | |
| Side-LargeModality=Side-view CXR, Model Type=Teacher Large2026.05 | 74.66 | 29.77 | — | — | 71.04 | 27.52 | |
| D-AllCategory=Tool Candidate Model2026.02 | 68.34 | 12.36 | 11.05 | 26.95 | 39.13 | — | |
| Hybrid LLMCategory=Tool Selection Method2026.02 | 66.13 | 8 | 13.8 | 19.06 | 35.58 | — | |
| D-PCCategory=Tool Candidate Model2026.02 | 60.22 | 13.35 | 9.72 | 42.39 | 38.74 | — | |
| RandomCategory=Tool Selection Method2026.02 | 60.15 | 9.54 | 7.56 | 27.19 | 32.2 | — | |
| D-NIHCategory=Tool Candidate Model2026.02 | 53.88 | 12.54 | 9.85 | 42.93 | 36.91 | — | |
| AutoMixCategory=Tool Selection Method2026.02 | 51.14 | 10.53 | 16.07 | 23.56 | 35.68 | — | |
| D-CXCategory=Tool Candidate Model2026.02 | 42.02 | 10.59 | 9.12 | 29.71 | 36.24 | — | |
| MF RouterCategory=Tool Selection Method2026.02 | 33.97 | 11.98 | 13.41 | 38.41 | 36.34 | — | |
| MLP RouterCategory=Tool Selection Method2026.02 | 33.69 | 9.01 | 11.46 | 31.92 | 34.29 | — | |
| SVM RouterCategory=Tool Selection Method2026.02 | 33.33 | 11.76 | 13.7 | 37.59 | 36.77 | — | |
| KNN RouterCategory=Tool Selection Method2026.02 | 30.16 | 11.52 | 11.63 | 38.75 | 35.09 | — | |
| E-T-CXCategory=Tool Candidate Model2026.02 | 28.63 | 9.12 | 9.54 | 37.83 | 35.67 | — | |
| E-S-CXCategory=Tool Candidate Model2026.02 | 28.53 | 11.44 | 11.5 | 39.91 | 35.43 | — | |
| Elo RouterCategory=Tool Selection Method2026.02 | 28.53 | 11.44 | 11.5 | 39.91 | 35.43 | — | |
| D-M-NBCategory=Tool Candidate Model2026.02 | 28.21 | 10.82 | 7.91 | 36.71 | 33.16 | — | |
| D-M-ChCategory=Tool Candidate Model2026.02 | 20.69 | 11.95 | 7.88 | 48.44 | 33.5 | — | |
| D-RSNACategory=Tool Candidate Model2026.02 | 12.32 | 11.07 | 8.84 | 54.16 | 32.15 | — | |
| Graph RouterCategory=Tool Selection Method2026.02 | 12.32 | 11.07 | 8.84 | 54.16 | 32.15 | — | |
| Causal LLMCategory=Tool Selection Method2026.02 | 12.32 | 11.07 | 8.84 | 54.16 | 32.15 | — |