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

Object Hallucination Evaluation on POPE (test)

90.6Accuracy

Vanilla

74.230478.480282.7386.9798May 30, 2024Sep 30, 2024Feb 1, 2025Jun 5, 2025Oct 7, 2025Feb 8, 2026Jun 12, 2026
Updated 1mo ago

Evaluation Results

MethodLinks
2025.12
90.6-----------100-------100--
90.6----------------------
2025.12
90.1-----------83.4-------61.9--
2025.12
89.6-----------82.7-------61.7--
2026.01
89.1----------------------
2026.01
89----------------------
2026.01
89----------------------
2026.01
88.9----------------------
2026.01
88.8----------------------
2026.01
88.8----------------------
2025.12
88.7-----------88.1-------63.9--
2026.01
88.7----------------------
2026.01
88.6----------------------
2026.05
88.6----------------------
2026.01
88.5----------------------
88.4----------------------
2026.05
88.4----------------------
2025.12
88.3-----------89.9-------64.4--
2026.01
88.2----------------------
88.2----------------------
2026.04
88.1----------------------
2026.04
87.9----------------------
2026.01
87.8----------------------
2026.04
87.6----------------------
2026.05
87.6----------------------
2026.04
87.5----------------------
2026.04
87.5----------------------
2026.05
87.5----------------------
2026.05
87.3----------------------
2024.05
87.2----------------------
2026.05
87.2----------------------
2026.04
87.1----------------------
2026.05
87.1----------------------
2026.04
87----------------------
2026.05
86.9----------------------
2026.04
86.8----------------------
2026.05
86.8----------------------
2026.05
86.8----------------------
86.7----------------------
2026.04
86.7----------------------
2026.05
86.7----------------------
86.4----------------------
2026.05
86.4----------------------
2026.05
86.4----------------------
2026.01
86.3----------------------
2026.05
86.3----------------------
2026.01
86.2----------------------
2026.04
86.2----------------------
2026.05
86.1----------------------
2026.04
86----------------------
2026.06
85.99----------------------
2024.05
85.9----------------------
2024.05
85.9----------------------
2026.02
85.9----------100-----------
2026.01
85.9----------------------
2026.06
85.84----------------------
2026.06
85.72----------------------
2026.05
85.7----------------------
2026.04
85.6----------------------
2026.05
85.4----------------------
2026.05
85.3----------------------
2026.05
85.3----------------------
2026.05
85.3----------------------
2026.04
85.2----------------------
2026.04
85----------------------
2026.01
84.8----------------------
2026.05
84.7----------------------
2026.05
84.7----------------------
2026.06
84.67----------------------
2026.04
84.5----------------------
2026.05
84.4----------------------
2026.04
84.3----------------------
2026.05
84.3----------------------
84.2----------------------
2026.05
84.2----------------------
2026.06
83.87----------------------
2026.05
83.8----------------------
2026.04
83.7----------------------
2026.04
83.5----------------------
2026.05
83.5----------------------
83.2----------------------
2025.12
82.7-----------78.6-------70.7--
2026.05
82.6----------------------
2026.06
82.45----------------------
2024.05
82.4----------------------
2026.04
81.9----------------------
2026.06
81.82----------------------
2026.04
81.1----------------------
2026.06
80.26----------------------
2024.05
80----------------------
2025.12
79.3-----------75.1-------73.4--
2024.05
78.9----------------------
2026.01
78.5----------------------
2026.06
77.79----------------------
2025.12
77.6-----------73.6-------71.5--
2026.02
77.5----------90.2-----------
2026.02
77----------89.6-----------
2026.02
75.7----------88.1-----------
2026.01
75.1----------------------
2026.06
74.86----------------------
Showing 100 of 215 rows