Hallucination Diagnosis on RSHBench
44.47Object Accuracy (OBJ)Qwen3-VL-4B
Evaluation Results
| Method | Links | ||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| Qwen3-VL-4BModel Category=Open-source Models2026.03 | 44.47 | 31.81 | 14.82 | 61.19 | 29.92 | 0.27 | 2.43 | 15.63 | 34.77 | 61.19 | |
| Claude-3-7Model Category=Closed-source Models2026.03 | 40.7 | 28.3 | 11.32 | 55.53 | 20.49 | 0.27 | 1.08 | 14.29 | 24.53 | 56.33 | |
| LLaMA-3.2-90BModel Category=Open-source Models2026.03 | 35.85 | 25.88 | 12.13 | 51.48 | 26.15 | 0.27 | 3.77 | 15.36 | 29.11 | 52.02 | |
| GLM-4.6vModel Category=Open-source Models2026.03 | 35.31 | 21.02 | 9.16 | 48.79 | 21.02 | 0 | 0.54 | 9.16 | 22.91 | 49.6 | |
| Gemini-2.5-proModel Category=Closed-source Models2026.03 | 33.42 | 31.54 | 15.36 | 48.79 | 27.49 | 0.54 | 0.81 | 15.09 | 29.65 | 49.06 | |
| GeoZeroModel Category=Open-source Models2026.03 | 33.42 | 33.96 | 15.9 | 49.87 | 28.3 | 3.77 | 2.16 | 18.06 | 29.65 | 49.87 | |
| GPT-4oModel Category=Closed-source Models2026.03 | 30.19 | 26.68 | 11.32 | 46.9 | 19.41 | 0.27 | 0.27 | 11.05 | 21.02 | 47.44 | |
| GeoZero+RADARModel Category=Open-source Models2026.03 | 28.03 | 25.61 | 13.48 | 38.54 | 21.83 | 2.16 | 1.89 | 15.63 | 24.8 | 38.81 | |
| LLaVA-1.5-7BModel Category=Open-source Models2026.03 | 25.88 | 29.11 | 14.29 | 46.9 | 25.61 | 2.96 | 2.7 | 16.71 | 26.95 | 47.71 |