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

Visual Reasoning on MMStar

69Accuracy

GazeVLM (Ours)

32.048841.641951.23560.8281Apr 30, 2025Jul 4, 2025Sep 8, 2025Nov 13, 2025Jan 17, 2026Mar 24, 2026May 29, 2026
Updated 2d ago

Evaluation Results

MethodLinks
2026.05
69--
2026.03
68.2--
2026.05
67.3--
2026.05
67.1--
2026.03
66.4--
2025.11
65.3121-
2026.05
65.07-1.34
2025.11
65175-
2025.11
64.9231-
2025.11
64.1246-
2025.11
63.9155-
2026.03
63.9--
2026.05
63.73--
2025.11
63.3200-
2026.03
61.5--
2026.05
61.4--
2026.05
61--
2026.03
60.9--
2026.03
60.7--
2026.05
60.67-0.6
2026.05
60.07--
2025.11
60221-
2026.03
59.9--
2026.05
59.27--
2026.03
58.6--
2026.05
58.3--
2026.05
57.9--
2025.11
57.6464-
2026.03
57.5--
2026.03
57.4--
2026.03
57.3--
2026.03
57--
2026.03
56.2--
2026.03
55.9--
2026.05
54.22--
2026.03
52.6--
2026.03
51.8--
2026.05
51.65--
2026.03
43.1--
2026.03
39.7--
2026.03
37.1--
2025.04
36.63--
2025.04
36.13--
2025.04
35.96--
2025.04
35.11--
2025.04
34.33--
2025.04
34.18--
2025.04
34.13--
2025.04
33.82--
2025.04
33.67--
2025.04
33.47--