Intervention on CogBiasESC (test)
3.12CogA (GPT)CoPoLLM-Llama3.1-8B
Evaluation Results
| Method | Links | ||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| CoPoLLM-Llama3.1-8BType=Ours2026.04 | 3.12 | 3.5 | 2.88 | 3.33 | 4.24 | 4.3 | 3.64 | 3.7 | 4.21 | 4.28 | 4.07 | 3.93 | |
| CoPoLLM-Qwen3-8BType=Ours2026.04 | 2.95 | 3.59 | 2.78 | 3.46 | 4.18 | 4.3 | 3.55 | 3.57 | 4.25 | 4.15 | 4.02 | 4.07 | |
| CoPoLLM-Qwen2.5-7BType=Ours2026.04 | 2.79 | 3.39 | 2.47 | 3.25 | 4.22 | 4.35 | 3.63 | 3.84 | 4.35 | 4.31 | 3.88 | 4 | |
| Grok-4-fastType=Close2026.04 | 2.53 | 2.96 | 2.26 | 2.75 | 4.2 | 4.44 | 3.67 | 3.72 | 4.32 | 4.18 | 3.67 | 3.83 | |
| Qwen2.5-7bType=Open2026.04 | 2.44 | 2.69 | 2.18 | 2.52 | 3.78 | 4.11 | 3.33 | 3.34 | 3.94 | 4.11 | 3.6 | 3.73 | |
| CPsyCounXType=Domain2026.04 | 2.4 | 2.61 | 2.1 | 2.26 | 3.26 | 3.34 | 3.14 | 2.97 | 3.08 | 2.99 | 3.49 | 2.99 | |
| Qwen-TurboType=Close2026.04 | 2.31 | 2.56 | 2.14 | 2.36 | 3.9 | 3.79 | 3.46 | 3.45 | 4.13 | 3.43 | 3.6 | 3.05 | |
| Mistral-7bType=Open2026.04 | 2.3 | 2.63 | 1.97 | 2.38 | 3.05 | 3.16 | 2.95 | 2.96 | 2.99 | 3.16 | 3.3 | 3.32 | |
| GPT4o-miniType=Close2026.04 | 2.27 | 2.45 | 2.09 | 2.09 | 3.85 | 3.97 | 3.33 | 3.22 | 3.99 | 3.95 | 3.5 | 3.58 | |
| XinjingType=Domain2026.04 | 2.13 | 2.25 | 1.96 | 1.95 | 3.54 | 3.04 | 3.45 | 2.92 | 3.2 | 3.1 | 3.44 | 2.98 | |
| Qwen3-8BType=Open2026.04 | 2.08 | 2.4 | 1.95 | 2.07 | 3.24 | 3.75 | 2.89 | 3.03 | 2.97 | 3.6 | 3.25 | 3.43 | |
| Gemini2.5-FlashType=Close2026.04 | 2.02 | 2.36 | 1.94 | 1.94 | 4.01 | 4.01 | 3.34 | 3.27 | 4.11 | 3.67 | 3.61 | 3.29 | |
| Llama3.1-8bType=Open2026.04 | 1.98 | 2.08 | 1.88 | 1.71 | 3.79 | 3.69 | 3.15 | 3.1 | 3.87 | 3.71 | 3.61 | 3.53 | |
| PsyDTLLM-Qwen2.5-7BType=Domain2026.04 | 1.97 | 2.35 | 1.87 | 1.98 | 3.73 | 3.9 | 3.2 | 3.3 | 3.82 | 3.82 | 3.5 | 3.43 | |
| PsycoLLMType=Domain2026.04 | 1.89 | 2.15 | 1.86 | 1.95 | 3.21 | 3.18 | 2.95 | 3.02 | 3.29 | 3.16 | 3.33 | 3.12 | |
| PsyDTLLM-Llama3.2-8BType=Domain2026.04 | 1.8 | 1.94 | 1.77 | 1.83 | 3.6 | 3.44 | 3.07 | 3.08 | 3.63 | 3.44 | 3.51 | 3.16 | |
| EmoLLMType=Domain2026.04 | 1.67 | 2.04 | 1.71 | 1.71 | 2.86 | 2.93 | 2.68 | 2.86 | 2.9 | 3.03 | 3.19 | 3.08 | |
| MindChatType=Domain2026.04 | 1.62 | 1.68 | 1.6 | 1.55 | 2.98 | 3.06 | 2.81 | 2.93 | 3.11 | 2.98 | 3.27 | 2.87 |