Multimodal Reward Modeling on PhyCritic-Bench
78.2Overall Score
Gemini-2.5-Pro
Updated 4d ago
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
2026.02 | 78.2 | 87.9 | 82.8 | 86.7 | 68.8 | 76.6 | - | |
2026.02 | 68 | 78.8 | 65.5 | 86.7 | 65.6 | 57.4 | - | |
2026.02 | 67.1 | 63.6 | 70.7 | 73.3 | 81.2 | 63.8 | - | |
2026.02 | 64.7 | 57.6 | 70.2 | 46.7 | 75 | 72.3 | - | |
2026.02 | 56 | 75.8 | 50 | 60 | 50 | 48.9 | - | |
2026.02 | 54.7 | 51.5 | 65.5 | 43.3 | 46.9 | 55.3 | - | |
2026.02 | 51.6 | 51.5 | 51.7 | 46.7 | 50 | 57.4 | - | |
2026.02 | 51.1 | 36.4 | 53.4 | 43.3 | 68.8 | 51.1 | - |