Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Visual Hallucination Mitigation on GPT-4 assisted benchmark
Loading...
46.7
SHR
Greedy
32.972
36.536
40.1
43.664
Jan 11, 2025
SHR
Updated 1mo ago
Evaluation Results
Method
Method
Links
SHR
Greedy
Backbone=MiniGPT-4
2025.01
46.7
VCD
Backbone=MiniGPT-4
2025.01
46
OPERA
Backbone=MiniGPT-4
2025.01
45.9
HALC
Backbone=MiniGPT-4
2025.01
45.8
VASparse
Backbone=MiniGPT-4
2025.01
45.2
Greedy
Backbone=mPLUG-Owl2
2025.01
42.3
VCD
Backbone=mPLUG-Owl2
2025.01
41.9
OPERA
Backbone=mPLUG-Owl2
2025.01
41.7
HALC
Backbone=mPLUG-Owl2
2025.01
41.7
VASparse
Backbone=mPLUG-Owl2
2025.01
41.1
Greedy
Backbone=LLaVA-1.5
2025.01
36.3
VCD
Backbone=LLaVA-1.5
2025.01
34.6
OPERA
Backbone=LLaVA-1.5
2025.01
34.2
HALC
Backbone=LLaVA-1.5
2025.01
33.9
VASparse
Backbone=LLaVA-1.5
2025.01
33.5
Feedback
Search any
task
Search any
task
