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

General Reasoning on MMBench

89.2Accuracy

GazeVLM (Ours)

80.77682.96385.1587.337May 8, 2026May 9, 2026May 10, 2026May 12, 2026May 13, 2026May 14, 2026May 16, 2026
Updated 15d ago

Evaluation Results

MethodLinks
2026.05
89.2
2026.05
88.7
2026.05
87.7
2026.05
84.6
2026.05
84
2026.05
84
2026.05
83.8
2026.05
83.8
2026.05
83.7
2026.05
83.4
2026.05
83.3
2026.05
82.9
2026.05
82.1
2026.05
82.1
2026.05
81.1