Visual Question Answering on VRSBench VQA 1.0 (test)
65.84Category AccuracyOurs
Evaluation results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| OursLLM=Vicuna-7B2025.12 | 65.84 | 90.23 | 64.26 | 51.82 | 59.18 | 64.48 | 87.42 | 62.68 | 45.91 | 65.76 | |
| GeoChatLLM=Vicuna-7B2025.12 | 62.27 | 86.67 | 73.58 | 32.65 | 49.05 | 26 | 42.16 | 22.07 | 44.03 | 56.59 | |
| SkySenseGPTLLM=Vicuna-7B2025.12 | 57.46 | 84.03 | 44.83 | 38 | 53.52 | 16.46 | 45.83 | 19.18 | 42.98 | 44.7 | |
| InternVL3.5LLM=InternLM2-8B2025.12 | 53.68 | 83.14 | 72.65 | 51.62 | 44.51 | 19.01 | 55.27 | 22.35 | 38.57 | 48.98 | |
| EarthDialLLM=Phi-3-mini-3.8B2025.12 | 51.49 | 47.22 | 36.21 | 41.33 | 36.62 | 11.39 | 50 | 22.07 | 35.43 | 36.86 | |
| VHMLLM=Vicuna-7B2025.12 | 50.75 | 86.81 | 36.21 | 42.67 | 53.52 | 55.7 | 54.17 | 51.51 | 32.29 | 51.51 | |
| MiniCPM-V-2.6LLM=Qwen2-7B2025.12 | 50.06 | 78.87 | 60.5 | 33.89 | 63.43 | 71.72 | 67.32 | 62.19 | 50.52 | 59.83 | |
| InternVL2.5LLM=Qwen2.5-7B2025.12 | 48.1 | 87.03 | 65.44 | 38.63 | 63.47 | 70.45 | 63.06 | 57.91 | 41.09 | 59.46 | |
| Gemini-2.0LLM=-2025.12 | 44.03 | 86.11 | 43.97 | 46 | 60.56 | 56.96 | 95.83 | 55.94 | 54.09 | 60.39 | |
| Claude-sonnet-4LLM=-2025.12 | 43.28 | 52.78 | 30.17 | 66.67 | 64.79 | 63.29 | 91.67 | 78.92 | 67.09 | 62.09 | |
| GLM-4.1V*LLM=GLM-4.1-9B, thinking version=true2025.12 | 42.54 | 86.11 | 43.1 | 54.67 | 69.01 | 62.03 | 87.5 | 45.68 | 56.81 | 60.83 | |
| Phi-3.5-VisionLLM=Phi-3-mini-3.8B2025.12 | 40.85 | 79.59 | 62.42 | 37.25 | 54.68 | 61.97 | 70.68 | 55.87 | 35.43 | 55.42 | |
| GPT-5LLM=-2025.12 | 39.55 | 88.89 | 42.24 | 47.33 | 70.42 | 59.49 | 87.5 | 62.33 | 42.98 | 60.08 | |
| Qwen2.5-VLLLM=Qwen2.5-7B2025.12 | 37.31 | 75.69 | 37.93 | 44 | 67.61 | 63.29 | 91.67 | 51.09 | 33.75 | 55.82 | |
| MiniGPT-v2LLM=LLaMA-2-7B2025.12 | 25.37 | 56.25 | 20.69 | 44 | 45.07 | 36.71 | 33.33 | 37.88 | 51.15 | 38.94 |