Multilingual Diagnostic Report Generation on Multilingual Diagnosis (15 Languages)
8.08Performance Score (ZH)MADE
Evaluation Results
| Method | Links | |||||||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MADEFramework=Multilingual Agentic Diagnosing Engine2026.06 | 8.08 | 7.34 | 7.72 | 7.86 | 7.23 | 8.26 | 8.31 | 7.94 | 8.06 | 7.87 | 8.25 | 8.3 | 8.25 | 8.29 | 8.47 | 8.02 | 100 | |
| opencodeFramework=General-purpose agent frameworks2026.06 | 5.86 | 4.87 | 4.89 | 5.33 | 5.09 | 5.49 | 5.43 | 5.63 | 5.71 | 5.77 | 5.56 | 5.35 | 5.54 | 5.45 | 5.63 | 5.45 | 97 | |
| nanobotFramework=General-purpose agent frameworks2026.06 | 5.59 | 4.64 | 4.68 | 4.92 | 4.87 | 5.3 | 5.42 | 5.32 | 5.27 | 4.74 | 5.57 | 5.36 | 5.49 | 4.84 | 5.23 | 5.15 | 99.6 | |
| smolagentsFramework=General-purpose agent frameworks2026.06 | 4.9 | 3.28 | 3.5 | 3.26 | 3.24 | 4.34 | 4.27 | 4.62 | 3.79 | 4.42 | 4.46 | 4.26 | 4.29 | 4.31 | 4.56 | 4.11 | 98.9 | |
| langchainFramework=General-purpose agent frameworks2026.06 | 4.63 | 3.41 | 3.4 | 3.67 | 3.68 | 4.05 | 4.01 | 4.18 | 3.98 | 3.77 | 4.1 | 3.94 | 4.06 | 3.73 | 4.28 | 3.93 | 99.3 | |
| genericagentFramework=General-purpose agent frameworks2026.06 | 4.22 | 3.26 | 2.89 | 3.25 | 3.14 | 3.7 | 3.44 | 3.79 | 3.59 | 3.55 | 3.57 | 3.44 | 3.74 | 3.45 | 3.46 | 3.5 | 100 | |
| cotFramework=Single-LLM2026.06 | 3.89 | 4 | 3.99 | 3.89 | 3.99 | 4.4 | 4.48 | 4.37 | 4.66 | 4.05 | 4.68 | 4.51 | 4.61 | 4.62 | 4.64 | 4.31 | 100 | |
| directFramework=Single-LLM2026.06 | 3.82 | 4.09 | 3.9 | 3.89 | 3.82 | 4.36 | 4.32 | 4.28 | 4.38 | 4.29 | 4.71 | 4.25 | 4.47 | 4.67 | 4.26 | 4.23 | 100 |