Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Object Hallucination Evaluation on POPE (dev)
Loading...
87
Accuracy
Qwen
72.752
76.451
80.15
83.849
Aug 1, 2025
Accuracy
Avg Relative Performance
Updated 1mo ago
Evaluation Results
Method
Method
Links
Accuracy
Avg Relative Performance
Qwen
Backbone=Qwen2.5-VL-3B...
2025.08
87
100
HiPrune
Backbone=Qwen2.5-VL-3B...
2025.08
85.9
98.7
HiPrune++
Backbone=Qwen2.5-VL-3B...
2025.08
85.9
98.7
VisionZip
Backbone=Qwen2.5-VL-3B...
2025.08
85.4
97.7
FastV
Backbone=Qwen2.5-VL-3B...
2025.08
85
97.4
HiPrune
Backbone=Qwen2.5-VL-3B...
2025.08
84.9
97.1
VisionZip
Backbone=Qwen2.5-VL-3B...
2025.08
84.6
96.2
HiPrune++
Backbone=Qwen2.5-VL-3B...
2025.08
84.4
97.1
FastV
Backbone=Qwen2.5-VL-3B...
2025.08
82.7
95.9
HiPrune
Backbone=Qwen2.5-VL-3B...
2025.08
80.4
93
VisionZip
Backbone=Qwen2.5-VL-3B...
2025.08
80.2
91.5
HiPrune++
Backbone=Qwen2.5-VL-3B...
2025.08
79.9
93
FastV
Backbone=Qwen2.5-VL-3B...
2025.08
73.3
86.4
Feedback
Search any
task
Search any
task
