Stylized Dialogue on 148-query style-1 persona (test)
4.257Context ScoreSFR
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| SFRTemperature=0.8, Backbone=Qwen3-32B, LLM-judge=Qwen3-235B2026.05 | 4.257 | 4.628 | 3.135 | 4.791 | |
| SFTTemperature=0.8, Backbone=Qwen3-32B, LLM-judge=Qwen3-235B2026.05 | 4.216 | 4.581 | 2.858 | 4.838 | |
| DeepSeek-R1-promptTemperature=0.8, Backbone=DeepSeek-R1 (frozen), LLM-judge=Qwen3-235B2026.05 | 4.175 | 4.562 | 3.117 | 4.781 |