Video Question Answering on Traffic-VQA (test)
83.156Overall Accuracy (OA)Qwen3-VL-8B + CTCNet
Evaluation Results
| Method | Links | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Qwen3-VL-8B + CTCNetModality=MUL2026.03 | 83.156 | 84.812 | 80.259 | 80.259 | 68.189 | 83.762 | 77.778 | 80.579 | 74.882 | |
| Qwen3-VL-8BModality=MUL, Variant=Finetuned, Venue=Ali 20252026.03 | 79.598 | 80.546 | 75.296 | 76.747 | 75.65 | 79.766 | 76.496 | 71.245 | 69.661 | |
| GPT-4oModality=MUL, Venue=OpenAI 20242026.03 | 67.72 | 75.28 | 73.96 | 77.67 | 73.13 | 74.06 | 69.51 | 78.03 | 69.13 | |
| Gemini-2.5-FlashModality=MUL, Venue=Google 20252026.03 | 67.47 | 74.92 | 73.65 | 75.91 | 71.37 | 75.39 | 75.99 | 79.2 | 69.76 | |
| GPT-4oModality=OPT, Venue=OpenAI 20242026.03 | 64.81 | 71.92 | 72.33 | 74.15 | 71.22 | 73.27 | 64.3 | 73.8 | 62.46 | |
| GPT-4oModality=TIR, Venue=OpenAI 20242026.03 | 63.66 | 69.97 | 70.67 | 71.13 | 66.74 | 71.98 | 65.72 | 76.06 | 59.23 | |
| CTCNetModality=MUL, Variant=Few-Shot2026.03 | 61.94 | 55.74 | 61.28 | 57.66 | 52.41 | 62.48 | 55.13 | 58.48 | 56.99 | |
| Gemini-2.5-ProModality=MUL, Venue=Google 20252026.03 | 61.41 | 70.05 | 64.93 | 69.38 | 66.03 | 68.64 | 68.72 | 72.99 | 62.15 | |
| Gemini-2.5-FlashModality=OPT, Venue=Google 20252026.03 | 59.51 | 66.7 | 69.77 | 65 | 66.8 | 73.34 | 63.35 | 70.22 | 62.17 | |
| Gemini-2.5-FlashModality=TIR, Venue=Google 20252026.03 | 57.01 | 61.53 | 61.58 | 57.55 | 59.04 | 69.24 | 64.3 | 71.39 | 51.13 | |
| Gemini-2.5-ProModality=OPT, Venue=Google 20252026.03 | 56.17 | 65.98 | 63.18 | 64.95 | 64.16 | 67.44 | 56.24 | 62.85 | 58.27 | |
| Gemini-2.5-ProModality=TIR, Venue=Google 20252026.03 | 51.5 | 57.51 | 55.14 | 53.18 | 55.17 | 64.76 | 47.87 | 61.39 | 46.98 | |
| Qwen3-VL-8BModality=MUL, Variant=Base, Venue=Ali 20252026.03 | 47.62 | 35.64 | 47.47 | 36.13 | 32.54 | 54.84 | 45.3 | 43.56 | 36.45 | |
| Qwen3-VL-8BModality=OPT, Variant=Base, Venue=Ali 20252026.03 | 47.56 | 35.11 | 46.79 | 34.76 | 32.67 | 56.01 | 31.45 | 40.02 | 42.95 | |
| CTCNetModality=OPT, Variant=Few-Shot2026.03 | 47.56 | 46.79 | 35.11 | 34.76 | 32.67 | 56.01 | 42.95 | 40.02 | 31.45 | |
| MiniCPM-VModality=MUL, Venue=arXiV 20242026.03 | 45.37 | 39.28 | 60.06 | 36.46 | 38.61 | 47.93 | 42.02 | 46.13 | 41.74 | |
| MiniCPM-VModality=OPT, Venue=arXiV 20242026.03 | 45.27 | 39.7 | 45.43 | 37.12 | 39.57 | 47.57 | 40.13 | 44.82 | 41.22 | |
| GeoChatModality=MUL, Venue=CVPR 20242026.03 | 44.22 | 41.66 | 51.19 | 41.8 | 41.65 | 45.27 | 43.92 | 43.87 | 47.19 | |
| Qwen3-VL-8BModality=TIR, Variant=Base, Venue=Ali 20252026.03 | 44.09 | 28.68 | 41.23 | 30.13 | 26.87 | 49.15 | 35.47 | 33.48 | 29.23 | |
| CTCNetModality=TIR, Variant=Few-Shot2026.03 | 44.09 | 41.23 | 28.68 | 30.13 | 26.87 | 49.15 | 35.47 | 33.48 | 29.23 | |
| GeoChatModality=OPT, Venue=CVPR 20242026.03 | 44 | 41.24 | 44.39 | 41.98 | 41.38 | 45.32 | 42.97 | 42.56 | 46.05 | |
| Qwen2.5-VL-7BModality=MUL, Venue=Ali 20252026.03 | 43.73 | 39.75 | 43.2 | 39.81 | 39.21 | 47.59 | 49.61 | 43.87 | 42.73 | |
| Qwen2.5-VL-7BModality=OPT, Venue=Ali 20252026.03 | 43.69 | 40.04 | 60.61 | 39.5 | 39.02 | 47.19 | 48.97 | 44.53 | 43.4 | |
| MiniCPM-VModality=TIR, Venue=arXiV 20242026.03 | 43.62 | 40.83 | 43.92 | 39.1 | 39.27 | 46.24 | 39.65 | 47.01 | 41.78 | |
| GeoPixModality=MUL, Venue=GRSM 20252026.03 | 43.62 | 38.11 | 51.42 | 37.21 | 39.9 | 45.12 | 37.91 | 43.21 | 41.17 | |
| GeoPixModality=OPT, Venue=GRSM 20252026.03 | 43.59 | 38.13 | 40.48 | 37.16 | 39.49 | 45.1 | 38.07 | 43.94 | 41.86 | |
| GeoChatModality=TIR, Venue=CVPR 20242026.03 | 43.08 | 40.42 | 45.72 | 39.85 | 41 | 47.77 | 45.81 | 42.85 | 42.35 | |
| GeoPixModality=TIR, Venue=GRSM 20252026.03 | 42.73 | 39.7 | 42.03 | 37.45 | 41.36 | 46.29 | 43.13 | 45.99 | 44.13 | |
| DeepSeek-VLModality=OPT, Venue=arXiV 20242026.03 | 42.64 | 38.62 | 41.51 | 37.01 | 37.79 | 44.38 | 41.23 | 42.92 | 34.86 | |
| DeepSeek-VLModality=MUL, Venue=arXiV 20242026.03 | 42.5 | 38.14 | 57.19 | 36.41 | 36.81 | 44.74 | 40.13 | 43.28 | 35.78 | |
| Qwen2.5-VL-7BModality=TIR, Venue=Ali 20252026.03 | 40.57 | 38.98 | 40.49 | 38.41 | 37.89 | 44.12 | 46.76 | 43.43 | 41.9 | |
| DeepSeek-VLModality=TIR, Venue=arXiV 20242026.03 | 39.2 | 34.63 | 38.39 | 33.27 | 34.46 | 39.09 | 36.34 | 39.71 | 33.93 | |
| MiniGPT-v2Modality=MUL, Venue=arXiv 20232026.03 | 36.68 | 35.91 | 37.69 | 34.2 | 38.11 | 39.23 | 31.75 | 39.2 | 42.54 | |
| MiniGPT-v2Modality=OPT, Venue=arXiv 20232026.03 | 36.04 | 35.3 | 35.95 | 33.74 | 37.38 | 37.96 | 29.54 | 39.05 | 40.14 | |
| FalconModality=MUL, Venue=arXiV 20252026.03 | 34.8 | 38.39 | 36.83 | 37.05 | 33.02 | 38.62 | 42.97 | 46.06 | 37.89 | |
| MiniGPT-v2Modality=TIR, Venue=arXiv 20232026.03 | 34.47 | 31.5 | 33.45 | 28.65 | 33.05 | 33.34 | 28.44 | 35.91 | 34.45 | |
| FalconModality=OPT, Venue=arXiV 20252026.03 | 34.22 | 38.27 | 40.41 | 37.57 | 32.45 | 37.83 | 43.29 | 46.42 | 37.84 | |
| FalconModality=TIR, Venue=arXiV 20252026.03 | 33.59 | 40.03 | 39.73 | 40.51 | 34.71 | 38.34 | 50.24 | 49.42 | 36.97 |