Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Multimodal Robustness on MMStar (test)

69.8MMStar Score

Qwen2.5-VL

49.41654.7086065.292Jan 15, 2026
Updated 5mo ago

Evaluation Results

MethodLinks
2026.01
69.8
2026.01
63.9
2026.01
59.4
2026.01
58.9
2026.01
55.53
2026.01
54.07
2026.01
53.73
2026.01
52.7
2026.01
52.1
2026.01
51.2
2026.01
50.2