Visual Question Answering on RealWorldQA xAI 2024 (test)
72.81AccuracyRobust-U1
Evaluation Results
| Method | Links | |
|---|---|---|
| Robust-U1Category=Ours, Intensity=clean2026.06 | 72.81 | |
| Robust-U1Category=Ours, Intensity=25%2026.06 | 72.81 | |
| Robust-U1Category=Ours, Intensity=50%2026.06 | 71.5 | |
| BAGELCategory=General MLLM, Intensity=clean2026.06 | 68.76 | |
| BAGELCategory=General MLLM, Intensity=50%2026.06 | 67.84 | |
| Robust-R1Category=Robust MLLM, Intensity=clean2026.06 | 67.71 | |
| Robust-U1Category=Ours, Intensity=100%2026.06 | 67.46 | |
| Robust-R1Category=Robust MLLM, Intensity=50%2026.06 | 67.05 | |
| Robust-R1Category=Robust MLLM, Intensity=25%2026.06 | 66.4 | |
| BAGELCategory=General MLLM, Intensity=25%2026.06 | 65.75 | |
| Qwen2.5-VL-3BCategory=General MLLM, Intensity=clean2026.06 | 65.22 | |
| Qwen2.5-VL-3BCategory=General MLLM, Intensity=25%2026.06 | 64.96 | |
| Qwen2.5-VL-3BCategory=General MLLM, Intensity=50%2026.06 | 63.39 | |
| Robust-R1Category=Robust MLLM, Intensity=100%2026.06 | 63.26 | |
| BAGELCategory=General MLLM, Intensity=100%2026.06 | 63.14 | |
| Qwen2.5-VL-3BCategory=General MLLM, Intensity=100%2026.06 | 60.65 | |
| InternVL-4BCategory=General MLLM, Intensity=25%2026.06 | 58.16 | |
| InternVL-4BCategory=General MLLM, Intensity=50%2026.06 | 57.64 | |
| InternVL-4BCategory=General MLLM, Intensity=clean2026.06 | 57.38 | |
| Gemma3-4BCategory=General MLLM, Intensity=clean2026.06 | 55.42 | |
| InternVL-4BCategory=General MLLM, Intensity=100%2026.06 | 54.9 | |
| Gemma3-4BCategory=General MLLM, Intensity=25%2026.06 | 54.77 | |
| Gemma3-4BCategory=General MLLM, Intensity=50%2026.06 | 53.72 | |
| Gemma3-4BCategory=General MLLM, Intensity=100%2026.06 | 52.81 | |
| Robust CLIPCategory=Robust MLLM, Intensity=clean2026.06 | 43.26 | |
| Robust CLIPCategory=Robust MLLM, Intensity=50%2026.06 | 42.61 | |
| Robust CLIPCategory=Robust MLLM, Intensity=25%2026.06 | 42.48 | |
| Robust CLIPCategory=Robust MLLM, Intensity=100%2026.06 | 41.43 | |
| TeCoACategory=Robust MLLM, Intensity=clean2026.06 | 40 | |
| TeCoACategory=Robust MLLM, Intensity=25%2026.06 | 39.73 | |
| TeCoACategory=Robust MLLM, Intensity=50%2026.06 | 39.47 | |
| TeCoACategory=Robust MLLM, Intensity=100%2026.06 | 38.69 |