Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Hallucination Evaluation on MSCOCO (val)
Loading...
6.87
Correctness
OPERA
5.1852
5.6226
6.06
6.4974
Nov 29, 2023
Correctness
Detailedness
Updated 3d ago
Evaluation Results
Method
Method
Links
Correctness
Detailedness
OPERA
Backbone Model=MiniGPT-4
2023.11
6.87
5.08
OPERA
Backbone Model=LLaVA-1.5
2023.11
6.32
5.16
OPERA
Backbone Model=Shikra
2023.11
6.29
5.26
OPERA
Backbone Model=Instruc...
2023.11
6.26
5.27
Beam Search
Backbone Model=LLaVA-1.5
2023.11
5.53
5.15
Beam Search
Backbone Model=Instruc...
2023.11
5.52
5.26
Beam Search
Backbone Model=MiniGPT-4
2023.11
5.29
5.06
Beam Search
Backbone Model=Shikra
2023.11
5.25
5.08
Feedback
Search any
task
Search any
task