Dialogue Generation on Cognitive stimulation simulated dataset 1.0 (test)
26.92ROUGE-LGCSD-3b
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| GCSD-3b2026.03 | 26.92 | 28.32 | 26.83 | 79.45 | 72.64 | 4.2 | 3.67 | 3.5 | |
| Doubao-Pro2026.03 | 23.58 | 22.39 | 21.51 | 73.19 | 71.03 | 4 | 3.55 | 3.33 | |
| GCSDCondition=w/o CSFAL2026.03 | 23.58 | 25.16 | 22.64 | 74.19 | 68.42 | 3.58 | 3.58 | 3.33 | |
| DeepSeek-671BPrompting=5-Shot2026.03 | 22.89 | 24.22 | 23.59 | 76.26 | 72.31 | 4.02 | 3.57 | 3.4 | |
| LLaMA-3.1-405B2026.03 | 22.56 | 22.13 | 21.08 | 76.52 | 71.26 | 3.75 | 3.7 | 3.42 | |
| GPT-4o2026.03 | 21.78 | 20.38 | 20.05 | 72.46 | 67.83 | 3.67 | 3.65 | 3.45 | |
| DeepSeek-671BPrompting=Zero-shot2026.03 | 21.67 | 22.13 | 22.09 | 77.29 | 73.79 | 3.59 | 3.45 | 3.36 | |
| ChatGPTPrompting=5-Shot2026.03 | 21.57 | 18.06 | 15.44 | 72.18 | 52.86 | 3.95 | 3.5 | 3.33 | |
| Qwen-MaxPrompting=5-Shot2026.03 | 21.54 | 21.34 | 20.15 | 74.27 | 64.41 | 3.65 | 3.6 | 3.38 | |
| GCSDCondition=w/o DPSM2026.03 | 21.53 | 24.45 | 22.18 | 79.33 | 56.41 | 3.6 | 3.54 | 3.32 | |
| GLM-4-PlusPrompting=5-Shot2026.03 | 17.69 | 18.24 | 18.11 | 70.18 | 72.97 | 3.45 | 3.15 | 3.15 | |
| ERNIE2026.03 | 17.24 | 8.32 | 7.38 | 65.35 | 53.92 | 3.05 | 3.2 | 2.9 |