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

Visual Grounded Reasoning on HR-Bench-4K

79.4Overall Score

Qwen2.5-VL-72B

63.69667.77371.8575.927Jul 10, 2025Aug 2, 2025Aug 25, 2025Sep 18, 2025Oct 11, 2025Nov 3, 2025Nov 27, 2025
Updated 1mo ago

Evaluation Results

MethodLinks
2025.11
79.488.870
2025.07
79.488.870
2025.11
78.691.865.4
2025.11
77.190.364
2025.07
77.190.364
2025.11
76.383.569
2025.11
75.584.566.5
2025.07
75.584.566.5
2025.11
75.191.359
2025.07
75.191.359
2025.11
74.889.360.3
2025.11
72.98660.3
2025.07
72.98660.3
2025.11
72.188.855.5
2025.07
72.188.855.5
2025.11
70.879.362.3
2025.07
70.879.362.3
2025.11
66.376.556
2025.07
66.376.556
2025.11
64.374.853.8
2025.07
64.374.853.8