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

Object Hallucination Evaluation on POPE GQA Popular

89.4Accuracy

SECOND

71.688876.286980.88585.4831Jun 10, 2025Jul 13, 2025Aug 16, 2025Sep 19, 2025Oct 22, 2025Nov 25, 2025Dec 29, 2025
Updated 19d ago

Evaluation Results

MethodLinks
2025.06
89.4-87.485.5
2025.06
87.8-87.283.3
2025.06
87.4-86.782.2
2025.12
86.8-87.3-
2025.06
86.8-86.383.1
2025.06
86.7-86.484.3
2025.06
86.6-86.384.8
2025.06
86.5-86.384.9
2025.06
85.3-87.592.1
2025.06
84.8-85.287.2
2025.06
84.6-84.986.3
2025.06
84.5-85.188
2025.12
84.4-86.2-
2025.06
84.2-84.888.2
2025.12
83.5487.2483.09-
2025.06
82.8-82.983.3
2025.06
82.5-81.476.6
2025.12
82.1384.5881.48-
2025.12
82.186.3981.85-
2025.12
81.9782.8281.73-
2025.12
81.3383.3880.74-
2025.12
81.1385.4881.03-
2025.06
81-79.774.5
2025.12
80.5-81.3-
2025.12
80.3-81.1-
2025.12
80.1-81.6-
2025.06
79.9-79.578.2
2025.12
79.5-79.8-
2025.12
78.5673.2883.34-
2025.12
78.2-77.1-
2025.12
78.1-78.3-
2025.12
77-80.1-
2025.12
76.8-78.6-
2025.12
76.7-78.7-
2025.12
76.1873.1778.65-
2025.12
76.1371.178.68-
2025.12
75.3471.8977.96-
2025.12
75.1769.9478.04-
2025.12
75.1271.5680.98-
2025.12
74.867.579.15-
2025.12
74.569.1777.61-
2025.12
73.8766.778.49-
2025.12
73.5-76.1-
2025.12
73.4766.8377.84-
2025.12
73.3368.7276.26-
2025.12
72.3765.2777.58-