Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Hallucination Evaluation on HallusionBench (test)
Loading...
17.8
Question Pair Accuracy
LPOI
7.972
10.5235
13.075
15.6265
May 27, 2025
Question Pair Accuracy
Figure Accuracy
Easy Accuracy
Hard Accuracy
Overall Accuracy
Updated 1mo ago
Evaluation Results
Method
Method
Links
Question Pair Accuracy
Figure Accuracy
Easy Accuracy
Hard Accuracy
Overall Accuracy
LPOI
Base Model=Idefics2-8B
2025.05
17.8
23.7
51.65
36.98
49.78
mDPO
Base Model=Idefics2-8B
2025.05
16.48
24.28
50.33
36.05
48.45
DPO
Base Model=Idefics2-8B
2025.05
15.82
22.54
49.45
33.72
46.68
Idefics2-8B
Base Model=Idefics2-8B
2025.05
8.35
14.16
32.53
30.93
35.08
Feedback
Search any
task
Search any
task