Visual Narrative Generation on MSBench Single-subject short-frame visual narratives
92.8CLIP-IIP-Adapter + 1Prompt1Story
Evaluation Results
| Method | Links | |||||
|---|---|---|---|---|---|---|
| IP-Adapter + 1Prompt1StoryResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 92.8 | 21.5 | 47.2 | 7.25 | 21.64 | |
| Qwen-EditResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 92.2 | 39.2 | 16.9 | — | — | |
| MS-Diffusion + 1Prompt1StoryResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 91.7 | 23.2 | 29.6 | 7.88 | 21.25 | |
| FluxKontextResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 91.3 | 38.1 | 18.7 | 32.31 | 22.99 | |
| Nano-BananaResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 91.1 | 38.9 | 18.3 | — | — | |
| MS-DiffusionResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 88.4 | 31.5 | 19.7 | 5.36 | 19.47 | |
| IP-AdapterResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 87.2 | 28.3 | 21.4 | 4.82 | 19.39 | |
| StoryTailorResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 86.9 | 43.1 | 17.1 | 7.02 | 20.01 | |
| λ-EclipseResolution=1024x1024, Sampling Steps=30, Hardware=RTX 40902026.02 | 82.6 | 29.9 | 26.8 | 8.31 | 14.21 |