User Study dataset
Benchmarks
| Task Name | Dataset Name | SOTA Result | Trend | |
|---|---|---|---|---|
| Personalized Image Generation | User Study Dataset (100 identity images: 50 synthetic StyleGAN3, 50 real Unsplash) 1.0 (test) | Baseline Selection Rate0.333 | 6 | |
| Human Preference Alignment | User study dataset HH-RLHF and PKU-SafeRLHF prompts (test) | Quality Score3.93 | 4 | |
| Motion Magnification | User Study Dataset fork, gunshot, flower (test) | Magnified Effect4.03 | 4 | |
| Image Perceptual Quality User Study | User Study dataset | Preference Rate90.8 | 3 | |
| User Preference Evaluation | User-study dataset SI tasks (Group 1) | Choice Count14 | 2 | |
| User Preference Evaluation | User-study dataset ST tasks (Group 1) | Choice Count16 | 2 | |
| User Preference Evaluation | User-study dataset CI tasks (Group 1) | Choice Count14 | 2 | |
| User Preference Evaluation | User-study dataset CT tasks (Group 1) | Choice Count13 | 2 | |
| Image Realism Classification | User Study dataset real vs. synthesized images | Accuracy65.5 | 2 |
Showing 9 of 9 rows