Human Preference Alignment on MM-AlignBench 1.0 (test)
84.9Win Rate
Claude3.5V-Sonnet
Updated 3d ago
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
2025.02 | 84.9 | 51.4 | 70 | 144 | 12 | 31 | 4 | |
2025.02 | 82.5 | 46 | 57 | 157 | 12 | 31 | 1 | |
2025.02 | 81.3 | 49 | 81 | 124 | 12 | 31 | 4 | |
2025.02 | 77 | 39.1 | 56 | 138 | 14 | 35 | 9 | |
2025.02 | 72.6 | 33.5 | 49 | 138 | 20 | 40 | 5 | |
2025.02 | 62.3 | 19.4 | 31 | 126 | 19 | 62 | 14 | |
2025.02 | 61.5 | 21.6 | 43 | 112 | 15 | 75 | 7 | |
2025.02 | 50 | 0 | - | - | - | - | - | |
2025.02 | 44.4 | -5.8 | 28 | 84 | 5 | 101 | 34 | |
2025.02 | 44.4 | -6.9 | 19 | 93 | 8 | 98 | 34 | |
2025.02 | 40.1 | -10.9 | 26 | 75 | 10 | 100 | 41 | |
2025.02 | 31.3 | -21.8 | 18 | 61 | 15 | 109 | 49 | |
2025.02 | 27.8 | -33.7 | 18 | 52 | 4 | 98 | 80 | |
2025.02 | 26.6 | -29 | 16 | 51 | 10 | 121 | 54 | |
2025.02 | 23.8 | -46.2 | 14 | 46 | 1 | 75 | 116 | |
2025.02 | 12.7 | -53 | 9 | 23 | 8 | 116 | 96 | |
2025.02 | 7.5 | -74 | 5 | 14 | 3 | 63 | 167 | |
2025.02 | 2.7 | -92.3 | 3 | 4 | 0 | 15 | 230 |