Psychiatric Diagnosis on PsySim-Adapt
72.9Depression PrecisionMIND-8B
Evaluation Results
| Method | Links | |||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| MIND-8BSetting=Fine-tuned2026.03 | 72.9 | 70 | 71.4 | 61.9 | 50.5 | 56.5 | 43.1 | 56 | 48.7 | 80 | 72 | 75.8 | 62.5 | |
| DDOSetting=Fine-tuned2026.03 | 59.5 | 53 | 56.1 | 53 | 50 | 51.5 | 46 | 42.5 | 44.2 | 52 | 46.5 | 49.1 | 45.7 | |
| MIND-4BSetting=Fine-tuned2026.03 | 59 | 53 | 56 | 52.5 | 50 | 51.3 | 45.5 | 42 | 43.9 | 53 | 48 | 50.4 | 47.5 | |
| DoctorAgent-RLSetting=Fine-tuned2026.03 | 58.5 | 53.5 | 55.9 | 52 | 51 | 51.5 | 38 | 40.5 | 39.2 | 52 | 47.5 | 49.6 | 47.8 | |
| GPT-4oSetting=Inference-only2026.03 | 58 | 52.5 | 55.2 | 41 | 37.8 | 39 | 28 | 20.5 | 22.2 | 6 | 5 | 5.5 | 40.5 | |
| Qwen3-8B†Setting=Fine-tuned2026.03 | 56.5 | 51 | 53.6 | 50 | 45 | 47.4 | 43 | 38 | 40.3 | 55 | 49 | 51.8 | 44 | |
| Qwen3-32BSetting=Inference-only2026.03 | 56 | 53.5 | 54.7 | 32 | 29 | 30.4 | 26 | 23 | 24.4 | 44 | 40.5 | 42 | 37.8 | |
| Qwen3-8BSetting=Inference-only2026.03 | 54 | 50.5 | 52 | 29 | 26 | 27.5 | 22 | 19 | 20.3 | 14 | 10.5 | 12 | 33.5 | |
| Baichuan-M2Setting=Inference-only2026.03 | 54 | 50.7 | 52.1 | 30 | 27 | 28.4 | 24 | 21 | 22.3 | 39 | 35.5 | 37 | 35.9 | |
| MRD-RAGSetting=Inference-only2026.03 | 53.5 | 49 | 51 | 44 | 41.5 | 42.9 | 32 | 29 | 30 | 23 | 20 | 21.4 | 36.5 | |
| DDTSetting=Inference-only2026.03 | 46.5 | 42.5 | 44.4 | 32 | 29.5 | 30.4 | 33 | 29.5 | 31 | 48 | 44 | 45.7 | 37 | |
| Qwen3-4BSetting=Inference-only2026.03 | 44 | 40.2 | 42 | 20 | 17 | 18.4 | 28 | 25 | 26.2 | 39 | 33.5 | 36 | 24.3 | |
| Qwen3-4B†Setting=Fine-tuned2026.03 | 44 | 38.5 | 41.1 | 14 | 10.5 | 12 | 24 | 22.5 | 23.1 | 38 | 33.5 | 35.6 | 27 | |
| DeepSeek-V3Setting=Inference-only2026.03 | 38 | 35 | 36.4 | 25 | 22 | 23.5 | 44 | 40 | 41.8 | 64 | 60.5 | 62.2 | 36.8 |