Pairwise Preference Prediction on DyCoBench-1K Overall Preference
78.3Preference Rate (A > B)DYCORM(OURS)
Evaluation results
| Method | Links | ||||
|---|---|---|---|---|---|
| DYCORM(OURS)Model Category=Reward Models2026.05 | 78.3 | 80.7 | 70.1 | 79.1 | |
| HPSV3Model Category=Reward Models2026.05 | 72.8 | 74.4 | 68.7 | 73.2 | |
| HPSV2Model Category=Reward Models2026.05 | 71.2 | 72.8 | 67.1 | 71.6 | |
| MPSModel Category=Reward Models2026.05 | 70.8 | 72.4 | 66.7 | 71.2 | |
| IMAGEREWARDModel Category=Reward Models2026.05 | 66.8 | 68.4 | 61.7 | 67.2 | |
| PICKSCOREModel Category=Reward Models2026.05 | 65.5 | 67.1 | 61.4 | 65.9 | |
| HPSModel Category=Reward Models2026.05 | 65.3 | 66.9 | 61.2 | 65.7 | |
| GPT-5 (25-08-07)Model Category=General LMMs2026.05 | 65.1 | 66.7 | 60 | 65.5 | |
| GEMINI-3.0-PROModel Category=General LMMs2026.05 | 64.7 | 66.3 | 61.3 | 65.1 | |
| GPT-4O (24-11-20)Model Category=General LMMs2026.05 | 64 | 65.6 | 57.9 | 64.4 | |
| AESTHETIC SCORE PREDICTORModel Category=Reward Models2026.05 | 63.5 | 65.1 | 58.4 | 63.9 | |
| CLIPSCOREModel Category=Reward Models2026.05 | 63.2 | 64.8 | 58.1 | 63.6 | |
| INTERNVL3.5-14BModel Category=General LMMs2026.05 | 61.2 | 62.8 | 55.1 | 61.6 | |
| QWEN3VL-8BModel Category=General LMMs2026.05 | 60.4 | 62 | 54.3 | 60.8 | |
| INTERNVL3.5-8BModel Category=General LMMs2026.05 | 59.8 | 61.4 | 53.7 | 60.2 | |
| LLAVA-ONEVISION-1.5-7BModel Category=General LMMs2026.05 | 59.6 | 61.2 | 53.5 | 60 | |
| QWEN3VL-4BModel Category=General LMMs2026.05 | 59.3 | 60.9 | 53.2 | 59.7 |