Human Evaluation on CulturalVQA OOD (test)
7.66FaithfulnessMMBoundary
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MMBoundary2025.05 | 7.66 | 7.17 | 8.26 | 7.69 | |
| SaySelf2025.05 | 7.12 | 6.81 | 6.58 | 6.83 | |
| RCE2025.05 | 6.84 | 6.19 | 6.82 | 6.62 | |
| DRL2025.05 | 6.55 | 5.63 | 6.07 | 6.08 | |
| Conf-CSR2025.05 | 6.38 | 5.45 | 5.86 | 5.89 | |
| Multisample2025.05 | 3.91 | 5.63 | 4.72 | 4.75 |