Free-format Multimodal Hallucination Assessment on RefoMB (dev)
62.9Trust
OmniLMM + RLAIF-V BoN
Updated 3d ago
Evaluation Results
| Method | Links | ||
|---|---|---|---|
2024.05 | 62.9 | 30.3 | |
2024.05 | 58.1 | 28.3 | |
2024.05 | 55.7 | 24.4 | |
2024.05 | 51.1 | 16.2 | |
2024.05 | 50 | 36.9 | |
2024.05 | 50 | 50 | |
2024.05 | 47.2 | 22.5 | |
2024.05 | 44.7 | 18.5 | |
2024.05 | 44.4 | 35.4 | |
2024.05 | 44.4 | 13.6 | |
2024.05 | 41.9 | 21.7 | |
2024.05 | 41.4 | 17.7 | |
2024.05 | 40.9 | 17.7 | |
2024.05 | 39.9 | 16.7 | |
2024.05 | 39.9 | 17.2 | |
2024.05 | 38.9 | 21.2 | |
2024.05 | 36.9 | 16.2 | |
2024.05 | 33.8 | 13.1 | |
2024.05 | 30.3 | 14.6 | |
2024.05 | 26.3 | 17.2 |