Content Quality Assessment on UltraVQA
69.7Acc@0.5ASO
Evaluation Results
| Method | Links | ||||
|---|---|---|---|---|---|
| ASOModel Category=Ours, Protocol=Analytic Score Optimization2026.02 | 69.7 | 0.52 | 0.796 | 0.737 | |
| GRPOModel Category=Ours, Protocol=Group Relative Policy Optimization2026.02 | 69 | 0.572 | 0.79 | 0.734 | |
| VideoScoreV2Model Category=VQA Baselines2026.02 | 60.2 | 0.818 | 0.689 | 0.628 | |
| FineVQModel Category=VQA Baselines2026.02 | 58.4 | 0.832 | 0.682 | 0.621 | |
| SFTModel Category=Ours, Protocol=Supervised Fine-Tuning2026.02 | 56.7 | 0.915 | 0.603 | 0.562 | |
| GPT-4.1Model Category=VLM-APIs2026.02 | 52.5 | 0.992 | 0.093 | 0.096 | |
| VideoLLaMA3Model Category=Open-source VLMs2026.02 | 51.8 | 1.006 | 0.205 | 0.157 | |
| MiniCPMV-4.5Model Category=Open-source VLMs2026.02 | 49.7 | 1.018 | 0.077 | 0.099 | |
| InternVideo-2.5Model Category=Open-source VLMs2026.02 | 46.1 | 1.041 | 0.028 | 0.051 | |
| Qwen2.5-VLModel Category=Open-source VLMs2026.02 | 43.3 | 1.22 | 0.093 | 0.097 | |
| Gemini-2.5ProModel Category=VLM-APIs2026.02 | 40.3 | 1.138 | 0.322 | 0.277 | |
| InternVL-3.5Model Category=Open-source VLMs2026.02 | 40.1 | 1.322 | 0.141 | 0.158 | |
| VideoChat-FlashModel Category=Open-source VLMs2026.02 | 29.6 | 1.38 | 0.188 | 0.183 | |
| Q-AlignModel Category=VQA Baselines2026.02 | 24.7 | 1.613 | 0.235 | 0.247 |