Narrative Generation on Creativityprism (test)
87.5NC (%)Vanilla
Evaluation results
| Method | Links | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| VanillaLLM Judge=DeepSeek-V32026.03 | 87.5 | 97.5 | 62.5 | 70 | 97.5 | 92.5 | 45 | 45 | 92.5 | 90 | 78 | |
| VanillaLLM Judge=Gemini-2.5-Flash2026.03 | 80 | 57.5 | 37.5 | 17.5 | 77.5 | 67.5 | 15 | 15 | 57.5 | 57.5 | 48.3 | |
| VanillaLLM Judge=GPT-5-Mini2026.03 | 75 | 75 | 50 | 45 | 75 | 75 | 25 | 25 | 70 | 62.5 | 57.2 | |
| Narrative StudioLLM Judge=GPT-5-Mini2026.03 | 15 | 5 | 35 | 32.5 | 20 | 20 | 50 | 50 | 20 | 25 | 27.8 | |
| Narrative StudioLLM Judge=Gemini-2.5-Flash2026.03 | 15 | 15 | 25 | 60 | 15 | 30 | 67.5 | 67.5 | 40 | 37.5 | 37.3 | |
| StoryWriterLLM Judge=DeepSeek-V32026.03 | 5 | 0 | 25 | 17.5 | 0 | 2.5 | 30 | 30 | 0 | 2.5 | 11.3 | |
| Narrative StudioLLM Judge=DeepSeek-V32026.03 | 5 | 0 | 7.5 | 12.5 | 0 | 2.5 | 22.5 | 22.5 | 0 | 2.5 | 7.3 | |
| StoryWriterLLM Judge=GPT-5-Mini2026.03 | 5 | 12.5 | 0 | 12.5 | 2.5 | 5 | 15 | 15 | 7.5 | 7.5 | 8.3 | |
| BiT-MCTSLLM Judge=GPT-5-Mini2026.03 | 5 | 7.5 | 15 | 10 | 2.5 | 0 | 10 | 10 | 2.5 | 5 | 6.8 | |
| StoryWriterLLM Judge=Gemini-2.5-Flash2026.03 | 5 | 15 | 20 | 22.5 | 2.5 | 2.5 | 12.5 | 12.5 | 2.5 | 5 | 10 | |
| BiT-MCTSLLM Judge=DeepSeek-V32026.03 | 2.5 | 2.5 | 5 | 0 | 2.5 | 2.5 | 2.5 | 2.5 | 7.5 | 5 | 3.3 | |
| BiT-MCTSLLM Judge=Gemini-2.5-Flash2026.03 | 0 | 12.5 | 17.5 | 0 | 5 | 0 | 5 | 5 | 0 | 0 | 4.5 |