Our new X account is live! Follow @wizwand_team for updates
WorkDL logo mark

Object Hallucination Probing on GQA POPE Popular

84.83Accuracy

SchroMind

56.084463.547271.0178.4728Mar 27, 2024Jul 19, 2024Nov 10, 2024Mar 4, 2025Jun 26, 2025Oct 18, 2025Feb 10, 2026
Updated 3d ago

Evaluation Results

MethodLinks
2026.02
84.83-----------84.93
2024.11
84.7-----------84.45
2024.11
84.7-----------84.78
2026.02
84.57-----------85.31
2024.11
84.23-----------83.92
2024.11
83.93-----------82.95
2024.11
82.63-----------82.22
2024.11
82.23-----------82.11
2024.11
81.97-----------82.12
2024.11
81.43-----------80.86
2026.02
80.86-----------82.64
2026.02
80.86-----------82.64
2024.11
80.73-----------82.24
2026.02
80.73-----------82.24
2026.02
80.73-----------82.24
2024.11
79.27-----------81.83
2026.02
79.27-----------82.11
2026.02
79.27-----------82.11
2024.03
78.8---------75.1587.5380.87
2024.03
78.57---------74.6286.6180.17
2024.11
78.17-----------80.73
2026.02
78.17-----------78.37
2026.02
78.17-----------78.37
2024.11
75.99-----------74.84
2026.02
74.8-----------79.17
2024.03
73.87---------60.6384.6976.42
2026.02
72.83-----------77.58
2024.03
62.84---------61.0980.5469.48
2024.03
62.14---------61.1472.2666.24
2026.01
58.87---------59.8753.856.67
2026.01
58.83---------58.6556.7357.68
2026.01
57.47---------57.2259.258.19
2024.03
57.19---------58.5560.8159.66
2025.06
-83.969.664.0364.5371.5363.4--66.62---
2025.06
--73.667.565.2371.8766.2--68.88---
2025.06
-81.5778.577.9379.680.57---79.63---
2025.06
-85.881.972.1775.07--59.270.471.93---
2025.06
--82.273.7374.17--54.776.8772.33---
2025.06
-84.4380.682.661.3--78.07-77.4---