Chat Generation on NoveltyBench
7.84LLM DivDiv Outline
Evaluation results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| Div OutlineBackbone LLM=Mistral-L3, Injection Level (Lv)=22026.06 | 7.84 | 46 | 26 | 15.74 | 7.91 | |
| KeywordBackbone LLM=Mistral-L3, Injection Level (Lv)=22026.06 | 7.33 | 42 | 23 | 15.13 | 8.16 | |
| Div OutlineBackbone LLM=DeepSeek-V3.2, Injection Level (Lv)=22026.06 | 7.13 | 50 | 32 | 18.04 | 7.43 | |
| KeywordBackbone LLM=DeepSeek-V3.2, Injection Level (Lv)=22026.06 | 6.77 | 47 | 29 | 17.41 | 7.77 | |
| DirectBackbone LLM=DeepSeek-V3.2, Injection Level (Lv)=0, Temperature (τ)=1.52026.06 | 3.59 | 43 | 25 | 19.28 | 6.9 | |
| DirectBackbone LLM=DeepSeek-V3.2, Injection Level (Lv)=02026.06 | 3.39 | 42 | 25 | 17.87 | 7.39 | |
| SSOTBackbone LLM=DeepSeek-V3.2, Injection Level (Lv)=12026.06 | 3.25 | 38 | 22 | 11.18 | 6.53 | |
| VerbalizedBackbone LLM=DeepSeek-V3.2, Injection Level (Lv)=02026.06 | 3.22 | 39 | 22 | 11.55 | 6.79 | |
| VerbalizedBackbone LLM=Mistral-L3, Injection Level (Lv)=02026.06 | 3.12 | 37 | 15 | 9.69 | 6.81 | |
| SSOTBackbone LLM=Mistral-L3, Injection Level (Lv)=12026.06 | 2.7 | 30 | 14 | 9.77 | 6.74 | |
| DirectBackbone LLM=Mistral-L3, Injection Level (Lv)=02026.06 | 2.59 | 45 | 16 | 14.48 | 6.8 |