Visual Question Answering on VG-150 (test)
55.03Recall (Category)AeroRAG
Evaluation results
| Method | Links | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| AeroRAG2026.04 | 55.03 | 34.68 | 13.08 | 11.72 | 49.73 | 30.56 | 14.43 | 4.59 | |
| Qwen-VL-Max2026.04 | 31.8 | 16.71 | 7.93 | 4.31 | 36.52 | 19.22 | 8.31 | 4.15 | |
| QvQ-72B2026.04 | 30.3 | 16.43 | 5.43 | 1.54 | 32.99 | 17.4 | 5.57 | 1.54 | |
| GPT-4o2026.04 | 30.17 | 19.36 | 7.34 | 4.04 | 35.43 | 21.98 | 8.6 | 3.94 | |
| Internvl2.52026.04 | 27.63 | 16.31 | 6.3 | 4.41 | 34.16 | 19.94 | 7.74 | 4.24 | |
| LAMM2026.04 | 24.75 | 17.25 | 0 | 0.59 | 26.96 | 17.85 | 0 | 0.88 | |
| Gemini Flash2026.04 | 23.55 | 14.43 | 5.83 | 2.52 | 32.22 | 19.93 | 7.53 | 3.25 |