Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Visual Understanding on RealWorldQA
Loading...
68.23
Accuracy (Clean)
Robust-R1 (SFT)
38.8708
46.4929
54.115
61.7371
Dec 19, 2025
Accuracy (Clean)
Accuracy (Intensity 25%)
Accuracy (Intensity 50%)
Accuracy (Intensity 100%)
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy (Clean)
Accuracy (Intensity 25%)
Accuracy (Intensity 50%)
Accuracy (Intensity 100%)
Robust-R1 (SFT)
Category=Ours, Trainin...
2025.12
68.23
67.58
67.32
63.92
Robust-R1 (SFT and RL)
Category=Ours, Trainin...
2025.12
67.71
66.4
67.05
63.26
Qwen2.5-VL-3B
Category=General MLLM
2025.12
65.22
64.96
63.39
60.65
InternVL-4B
Category=General MLLM
2025.12
57.38
58.16
57.64
54.9
Gemma3-4B
Category=General MLLM
2025.12
55.42
54.77
53.72
52.81
Robust CLIP
Category=Robust MLLM
2025.12
43.26
42.48
42.61
41.43
TeCoA
Category=Robust MLLM
2025.12
40
39.73
39.47
38.69
Feedback
Search any
task
Search any
task