Visual Question Answering on MMMB (test)
84.75AccuracyRobust-U1
Evaluation Results
| Method | Links | |
|---|---|---|
| Robust-U1Category=Ours, Intensity=clean2026.06 | 84.75 | |
| Robust-U1Category=Ours, Intensity=25%2026.06 | 84.14 | |
| Robust-U1Category=Ours, Intensity=50%2026.06 | 83.54 | |
| Robust-U1Category=Ours, Intensity=100%2026.06 | 83.18 | |
| BAGELCategory=General MLLM, Intensity=clean2026.06 | 81.92 | |
| Robust-R1Category=Robust MLLM, Intensity=clean2026.06 | 81.41 | |
| BAGELCategory=General MLLM, Intensity=25%2026.06 | 81.16 | |
| Qwen2.5-VL-3BCategory=General MLLM, Intensity=clean2026.06 | 80.6 | |
| BAGELCategory=General MLLM, Intensity=50%2026.06 | 80.56 | |
| Robust-R1Category=Robust MLLM, Intensity=25%2026.06 | 79.49 | |
| Qwen2.5-VL-3BCategory=General MLLM, Intensity=25%2026.06 | 79.19 | |
| Robust-R1Category=Robust MLLM, Intensity=50%2026.06 | 79.04 | |
| Qwen2.5-VL-3BCategory=General MLLM, Intensity=50%2026.06 | 78.68 | |
| BAGELCategory=General MLLM, Intensity=100%2026.06 | 78.48 | |
| InternVL-4BCategory=General MLLM, Intensity=clean2026.06 | 77.97 | |
| InternVL-4BCategory=General MLLM, Intensity=25%2026.06 | 77.47 | |
| InternVL-4BCategory=General MLLM, Intensity=50%2026.06 | 76.66 | |
| Robust-R1Category=Robust MLLM, Intensity=100%2026.06 | 75.35 | |
| InternVL-4BCategory=General MLLM, Intensity=100%2026.06 | 74.59 | |
| Qwen2.5-VL-3BCategory=General MLLM, Intensity=100%2026.06 | 74.5 | |
| Gemma3-4BCategory=General MLLM, Intensity=clean2026.06 | 71.01 | |
| Gemma3-4BCategory=General MLLM, Intensity=25%2026.06 | 70.3 | |
| Gemma3-4BCategory=General MLLM, Intensity=50%2026.06 | 70.2 | |
| Gemma3-4BCategory=General MLLM, Intensity=100%2026.06 | 69.14 | |
| TeCoACategory=Robust MLLM, Intensity=25%2026.06 | 65.71 | |
| Robust CLIPCategory=Robust MLLM, Intensity=clean2026.06 | 58.83 | |
| Robust CLIPCategory=Robust MLLM, Intensity=25%2026.06 | 58.28 | |
| Robust CLIPCategory=Robust MLLM, Intensity=50%2026.06 | 57.97 | |
| TeCoACategory=Robust MLLM, Intensity=clean2026.06 | 57.17 | |
| TeCoACategory=Robust MLLM, Intensity=50%2026.06 | 56.11 | |
| Robust CLIPCategory=Robust MLLM, Intensity=100%2026.06 | 53.33 | |
| TeCoACategory=Robust MLLM, Intensity=100%2026.06 | 51.76 |