Depression Severity Assessment on DepSeverity D3 (test)
80.81AccuracyMental-R1
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Mental-R1Architecture=CRPO-trained2026.06 | 80.81 | 78.86 | |
| GPT-5Category=Industry-level models2026.06 | 63.22 | 62.19 | |
| DeepSeek-R1Category=Industry-level models2026.06 | 62.51 | 62.57 | |
| Llama-3.1-SFTCategory=Domain and public models, Training Mode=SFT2026.06 | 58.76 | 63.67 | |
| GPT-4oCategory=Industry-level models2026.06 | 58.34 | 52.74 | |
| Gemma-2-SFTCategory=Domain and public models, Training Mode=SFT2026.06 | 57.24 | 56.01 | |
| GPT-3.5Category=Industry-level models2026.06 | 53.41 | 50.47 | |
| Mental-GLMCategory=Domain and public models2026.06 | 53.25 | 53.63 | |
| DeepSeek-V3Category=Industry-level models2026.06 | 47.41 | 49.14 | |
| MentallamaCategory=Domain and public models2026.06 | 45.15 | 41.95 |