Sports Video Question Answering on SportsTime
39.66Perception ScoreGemini-2.5-Pro
Evaluation Results
| Method | Links | ||||||
|---|---|---|---|---|---|---|---|
| Gemini-2.5-ProVisual Input=0.2 fps2026.04 | 39.66 | 19.78 | 29.49 | 39.41 | 13.71 | 29.37 | |
| GPT-5Visual Input=0.2 fps2026.04 | 38.77 | 27.45 | 43.85 | 44.77 | 46.93 | 40.72 | |
| CoTR (full)training=full, Backbone=Qwen3-VL-4B-Instruct2026.04 | 25.12 | 14.93 | 34.56 | 26.12 | 40.21 | 29.23 | |
| Qwen3-VL-8B-InstructVisual Input=7682026.04 | 24.27 | 14.31 | 33.26 | 23.08 | 44.47 | 27.45 | |
| Qwen3-VL-4B-InstructParameters=4B2026.04 | 23.26 | 13.14 | 31.11 | 22.92 | 34.17 | 25.15 | |
| Qwen3-VL-4B-InstructVisual Input=7682026.04 | 23.26 | 13.14 | 31.11 | 22.92 | 34.17 | 25.15 | |
| CoTR (tr-GRPO)training=tr-GRPO, Backbone=Qwen3-VL-4B-Instruct2026.04 | 22.01 | 13.73 | 35.73 | 25.54 | 39.81 | 27.31 | |
| InternVL2.5-4BParameters=4B2026.04 | 21.99 | 14.83 | 30.58 | 16.3 | 39.23 | 24.57 | |
| VideoLLaMA-7BVisual Input=7682026.04 | 20.06 | 10.98 | 16.92 | 14 | 28.54 | 17.96 | |
| Video-R1-7BVisual Input=7682026.04 | 19.26 | 8.04 | 23.25 | 18.62 | 33.01 | 20.4 | |
| InternVideo2.5-8BVisual Input=5122026.04 | 18.12 | 10.39 | 19.83 | 9.85 | 26.21 | 16.68 | |
| Qwen2.5-VL-3B-InstructParameters=3B2026.04 | 16.67 | 7.06 | 27.18 | 12.31 | 22.52 | 17.16 | |
| MiniCPM-V4.5-8BVisual Input=5122026.04 | 16.38 | 10.14 | 13.5 | 16.18 | 27.27 | 16.48 | |
| MiniCPM-v4-4BParameters=4B2026.04 | 14.21 | 10.36 | 19.2 | 12.08 | 26.06 | 16.23 | |
| LLaVA-OneVision1.5-4BParameters=4B2026.04 | 13.88 | 7.4 | 13.61 | 7.73 | 16.57 | 11.81 | |
| Ovis2-4BParameters=4B2026.04 | 11.87 | 6.34 | 13.96 | 6.76 | 5.25 | 9.02 | |
| GLM-4.6v-Flash-9BVisual Input=5122026.04 | 2.27 | 1.18 | 3.76 | 3.54 | 13.2 | 4.62 |