Cognitive Distortion Diagnose on CogBiasESC (test)
72.6PrecisionCoPoLLM-Qwen2.5-7B
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| CoPoLLM-Qwen2.5-7BType=Ours2026.04 | 72.6 | 50.7 | 59.7 | 40.7 | |
| CoPoLLM-Qwen3-8BType=Ours2026.04 | 64.7 | 63.6 | 64.1 | 30.5 | |
| Gemini2.5-FlashType=Close2026.04 | 59.9 | 56.2 | 58 | 57.6 | |
| CoPoLLM-Llama3.1-8BType=Ours2026.04 | 57.8 | 60.4 | 59.1 | 20.3 | |
| Grok-4-fastType=Close2026.04 | 54 | 55.5 | 54.7 | 55.9 | |
| Qwen-TurboType=Close2026.04 | 46 | 57.9 | 51.3 | 50.9 | |
| EmoLLMType=Domain2026.04 | 43.7 | 50.8 | 47 | 39 | |
| GPT4o-miniType=Close2026.04 | 43 | 50 | 46.2 | 40.7 | |
| Qwen3-8BType=Open2026.04 | 38.1 | 44.5 | 41.1 | 44.1 | |
| Qwen2.5-7bType=Open2026.04 | 37.7 | 36.7 | 37.2 | 79.7 | |
| PsycoLLMType=Domain2026.04 | 35.7 | 50.3 | 41.8 | 86.4 | |
| Llama3.1-8bType=Open2026.04 | 30.7 | 51.5 | 38.5 | 44.1 | |
| Mistral-7bType=Open2026.04 | 28.3 | 17.2 | 21.4 | 98.3 | |
| PsyDTLLM-Qwen2.5-7BType=Domain2026.04 | 27.6 | 44.4 | 34 | 55.9 | |
| CPsyCounXType=Domain2026.04 | 26.4 | 29.3 | 27.8 | 83.1 | |
| MindChatType=Domain2026.04 | 24.7 | 27.6 | 26.1 | 74.6 | |
| PsyDTLLM-Llama3.2-8BType=Domain2026.04 | 23.3 | 36.3 | 28.3 | 61.4 | |
| XinjingType=Domain2026.04 | 23.1 | 55.7 | 30.1 | 45.8 |