Human Evaluation on A-OKVQA (test)
7.83Faithfulness ScoreMMBoundary
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| MMBoundary2025.05 | 7.83 | 7.25 | 8.18 | 7.75 | |
| SaySelf2025.05 | 7.28 | 7.49 | 6.47 | 7.08 | |
| RCE2025.05 | 6.73 | 6.58 | 7.41 | 6.9 | |
| DRL2025.05 | 6.54 | 6.13 | 6.95 | 6.54 | |
| Conf-CSR2025.05 | 6.47 | 5.73 | 5.82 | 6.01 | |
| Multisample2025.05 | 4.2 | 5.17 | 4.06 | 4.47 |