List your company

Visual Question Answering on MMMB (test)

84.75AccuracyRobust-U1
50.440459.347768.25577.1623Jun 6, 2026

Evaluation results

MethodLinks
Robust-U1Category=Ours, Intensity=clean2026.0684.75
Robust-U1Category=Ours, Intensity=25%2026.0684.14
Robust-U1Category=Ours, Intensity=50%2026.0683.54
Robust-U1Category=Ours, Intensity=100%2026.0683.18
BAGELCategory=General MLLM, Intensity=clean2026.0681.92
Robust-R1Category=Robust MLLM, Intensity=clean2026.0681.41
BAGELCategory=General MLLM, Intensity=25%2026.0681.16
Qwen2.5-VL-3BCategory=General MLLM, Intensity=clean2026.0680.6
BAGELCategory=General MLLM, Intensity=50%2026.0680.56
Robust-R1Category=Robust MLLM, Intensity=25%2026.0679.49
Qwen2.5-VL-3BCategory=General MLLM, Intensity=25%2026.0679.19
Robust-R1Category=Robust MLLM, Intensity=50%2026.0679.04
Qwen2.5-VL-3BCategory=General MLLM, Intensity=50%2026.0678.68
BAGELCategory=General MLLM, Intensity=100%2026.0678.48
InternVL-4BCategory=General MLLM, Intensity=clean2026.0677.97
InternVL-4BCategory=General MLLM, Intensity=25%2026.0677.47
InternVL-4BCategory=General MLLM, Intensity=50%2026.0676.66
Robust-R1Category=Robust MLLM, Intensity=100%2026.0675.35
InternVL-4BCategory=General MLLM, Intensity=100%2026.0674.59
Qwen2.5-VL-3BCategory=General MLLM, Intensity=100%2026.0674.5
Gemma3-4BCategory=General MLLM, Intensity=clean2026.0671.01
Gemma3-4BCategory=General MLLM, Intensity=25%2026.0670.3
Gemma3-4BCategory=General MLLM, Intensity=50%2026.0670.2
Gemma3-4BCategory=General MLLM, Intensity=100%2026.0669.14
TeCoACategory=Robust MLLM, Intensity=25%2026.0665.71
Robust CLIPCategory=Robust MLLM, Intensity=clean2026.0658.83
Robust CLIPCategory=Robust MLLM, Intensity=25%2026.0658.28
Robust CLIPCategory=Robust MLLM, Intensity=50%2026.0657.97
TeCoACategory=Robust MLLM, Intensity=clean2026.0657.17
TeCoACategory=Robust MLLM, Intensity=50%2026.0656.11
Robust CLIPCategory=Robust MLLM, Intensity=100%2026.0653.33
TeCoACategory=Robust MLLM, Intensity=100%2026.0651.76