Psychological defense mechanism classification on PSYDEFCONV (test)
64.83AccuracyMinistral-8B
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| Ministral-8BEvaluation Protocol=Fine-tuned2025.12 | 64.83 | 33.97 | 30.45 | 31.48 | |
| InternLM3-8BEvaluation Protocol=Fine-tuned2025.12 | 63.98 | 33.47 | 29.93 | 30.53 | |
| Llama3.1-8BEvaluation Protocol=Fine-tuned2025.12 | 62.92 | 33.24 | 30.08 | 30.51 | |
| GLM-4-9BEvaluation Protocol=Fine-tuned2025.12 | 62.92 | 30.1 | 29.53 | 28.61 | |
| Qwen3-8BEvaluation Protocol=Fine-tuned2025.12 | 61.44 | 30.1 | 28.91 | 28.39 | |
| Qwen3-4BEvaluation Protocol=Fine-tuned2025.12 | 60.59 | 30.49 | 28.53 | 28.46 | |
| Gemini 2.5 ProEvaluation Protocol=Zero-shot Prompted2025.12 | 56.36 | 27.49 | 26.12 | 25.99 | |
| DeepSeek-V3.2Evaluation Protocol=Zero-shot Prompted, Thinking Process=enabled2025.12 | 55.72 | 29.66 | 27.53 | 26.17 | |
| GPT-5 miniEvaluation Protocol=Zero-shot Prompted2025.12 | 54.03 | 26.3 | 16.99 | 18.41 | |
| GPT-5Evaluation Protocol=Zero-shot Prompted2025.12 | 52.75 | 27.59 | 16.56 | 19.53 | |
| Qwen3-NextEvaluation Protocol=Zero-shot Prompted, Thinking Process=enabled2025.12 | 44.28 | 27.23 | 21.58 | 20.68 | |
| Kimi-K2Evaluation Protocol=Zero-shot Prompted2025.12 | 41.1 | 21.28 | 21.38 | 17.58 | |
| Qwen3-NextEvaluation Protocol=Zero-shot Prompted, Thinking Process=disabled2025.12 | 40.89 | 26.91 | 19.09 | 14.88 | |
| DeepSeek-V3.2Evaluation Protocol=Zero-shot Prompted, Thinking Process=disabled2025.12 | 39.83 | 24.77 | 19.97 | 16.15 |