Story Generation on Story Generation Evaluation Set (Subjective Metrics)
89.45CreativityStorytelling
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| StorytellingStory Type=Storytelling, Model=Gemma2025.10 | 89.45 | 92.15 | 92.75 | 85.65 | 96.05 | 91.21 | |
| StorytellingStory Type=Storytelling, Model=Llama32025.10 | 79.5 | 94.75 | 89.45 | 85.65 | 96.85 | 89.24 | |
| Storytelling w/o Role-PlayingStory Type=w/o Role-Playing, Model=Gemma2025.10 | 79.45 | 86.8 | 83.95 | 83.15 | 91.05 | 84.88 | |
| BaselineStory Type=Baseline, Model=Gemma2025.10 | 65.25 | 68.3 | 80.15 | 71.2 | 78.9 | 72.76 | |
| StorytellingStory Type=Storytelling, Model=GPT4o2025.10 | 63.15 | 63.45 | 59.35 | 70.9 | 69.1 | 65.19 | |
| BaselineStory Type=Baseline, Model=Llama32025.10 | 59.25 | 71.55 | 76.15 | 71.6 | 70 | 69.71 | |
| Storytelling w/o Environment TrajectoriesStory Type=w/o Environment Trajectories, Model=Gemma2025.10 | 55.3 | 74.35 | 78.8 | 73.45 | 85.5 | 73.48 | |
| BaselineStory Type=Baseline, Model=GPT4o2025.10 | 50 | 50 | 50 | 50 | 50 | 50 | |
| Storytelling w/o Role-PlayingStory Type=w/o Role-Playing, Model=Llama32025.10 | 49.35 | 73.65 | 72 | 73.7 | 74.35 | 68.61 | |
| Storytelling w/o Environment TrajectoriesStory Type=w/o Environment Trajectories, Model=Llama32025.10 | 31.2 | 52.7 | 39.2 | 51.75 | 50.85 | 45.14 | |
| Storytelling w/o Environment TrajectoriesStory Type=w/o Environment Trajectories, Model=GPT4o2025.10 | 24.35 | 21.2 | 26.6 | 21.05 | 18.85 | 22.41 | |
| Storytelling w/o Role-PlayingStory Type=w/o Role-Playing, Model=GPT4o2025.10 | 18.05 | 45.8 | 47.9 | 49.7 | 48.3 | 41.95 |