Psychiatric Diagnosis on PsySim Std
79.6Depression PrecisionMIND-8B
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MIND-8BSetting=Fine-tuned2026.03 | 79.6 | 86 | 80.7 | 100 | 66.1 | 73.3 | 48.8 | 78 | 69.2 | 86.8 | 92 | 78.3 | 71.5 | |
| DDOSetting=Fine-tuned2026.03 | 73 | 66.1 | 69.4 | 69.1 | 66 | 67.5 | 58.8 | 55.4 | 57.1 | 66.7 | 59 | 62.6 | 53 | |
| DoctorAgent-RLSetting=Fine-tuned2026.03 | 71.6 | 66.9 | 69.2 | 70.2 | 64.8 | 67.4 | 60.7 | 57.6 | 59.1 | 64.3 | 60.8 | 62.5 | 56.5 | |
| Qwen3-8B†Setting=Fine-tuned2026.03 | 69.2 | 63.4 | 66.1 | 73.5 | 65.8 | 69.4 | 62.1 | 57.8 | 59.9 | 68 | 63.2 | 65.5 | 53.5 | |
| GPT-4oSetting=Inference-only2026.03 | 66 | 60.5 | 63.1 | 49 | 45.6 | 47.2 | 42 | 38.2 | 40 | 12.5 | 10.3 | 11.3 | 49.5 | |
| Qwen3-32BSetting=Inference-only2026.03 | 64 | 61.1 | 62.5 | 40 | 36.2 | 38 | 34 | 30.2 | 32 | 52 | 48.2 | 50 | 47.2 | |
| Qwen3-8BSetting=Inference-only2026.03 | 62 | 58.2 | 60 | 37 | 33.3 | 35 | 30 | 26.3 | 28 | 22 | 18.4 | 20 | 43 | |
| Baichuan-M2Setting=Inference-only2026.03 | 62 | 58.3 | 60.1 | 38 | 34.3 | 36.1 | 32 | 28.3 | 30 | 47 | 43.3 | 45.1 | 44.7 | |
| MIND-4BSetting=Fine-tuned2026.03 | 62 | 65 | 63.5 | 54 | 45 | 49.1 | 46 | 40 | 42.8 | 58 | 64 | 60.9 | 54.5 | |
| MRD-RAGSetting=Inference-only2026.03 | 61.5 | 56.8 | 59.1 | 52 | 49.2 | 50.6 | 40 | 36.9 | 38.4 | 31 | 28 | 29.4 | 46 | |
| Qwen3-4B†Setting=Fine-tuned2026.03 | 60 | 54 | 56.8 | 28 | 22 | 24.6 | 30 | 28 | 28.9 | 52 | 48 | 49.9 | 35.5 | |
| DDTSetting=Inference-only2026.03 | 54.5 | 50.7 | 52.5 | 40 | 36.9 | 38.4 | 41 | 36.9 | 38.8 | 56 | 51.9 | 53.9 | 46.5 | |
| Qwen3-4BSetting=Inference-only2026.03 | 52 | 48.2 | 50 | 28 | 24.3 | 26 | 14 | 10.6 | 12 | 47 | 41.5 | 44.1 | 32.8 | |
| DeepSeek-V3Setting=Inference-only2026.03 | 46 | 43 | 44.4 | 33 | 30 | 31.4 | 52 | 47.7 | 49.8 | 72 | 68.4 | 70.2 | 46 |