Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Generative Hallucination Mitigation on MMHal-Bench
Loading...
3.49
Overall Score
GPT-4V
1.4724
1.9962
2.52
3.0438
Apr 20, 2026
Overall Score
Hallucination Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Overall Score
Hallucination Rate
GPT-4V
Paradigm=Standard LVLM
2026.04
3.49
28
PSRD
Paradigm=Post-hoc hall...
2026.04
2.92
49
HSA-DPO
Paradigm=Fine-tune LVL...
2026.04
2.61
48
Octopus
Paradigm=Post-hoc hall...
2026.04
2.61
50
LLaVA-DPO
Paradigm=Fine-tune LVL...
2026.04
2.58
50
SENA
Paradigm=Fine-tune LVL...
2026.04
2.33
52
AVISC
Paradigm=Post-hoc hall...
2026.04
2.19
59
OPERA
Paradigm=Post-hoc hall...
2026.04
2.15
54
M3ID
Paradigm=Post-hoc hall...
2026.04
2.14
61
STIC
Paradigm=Fine-tune LVL...
2026.04
2.07
56
EOS
Paradigm=Fine-tune LVL...
2026.04
2.03
59
VCD
Paradigm=Post-hoc hall...
2026.04
1.96
64
LLaVA-1.5-7B (Base)
Paradigm=Standard LVLM
2026.04
1.55
76
Feedback
Search any
task
Search any
task