Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Vision-Language Hallucination Assessment on MM-HallBench
Loading...
3.64
Average Score
Greedy
3.3072
3.3936
3.48
3.5664
Apr 25, 2026
Average Score
Hallucination Rate
Updated 1mo ago
Evaluation Results
Method
Method
Links
Average Score
Hallucination Rate
Greedy
Backbone=Qwen2.5-VL (3B)
2026.04
3.64
34
Top-p (nucleus)
Backbone=Qwen2.5-VL (3B)
2026.04
3.58
36
η-Sampling
Backbone=Qwen2.5-VL (3B)
2026.04
3.57
38
Beam Search
Backbone=Qwen2.5-VL (3B)
2026.04
3.54
34
Temp’ Only
Backbone=Qwen2.5-VL (3B)
2026.04
3.48
36
Top-k
Backbone=Qwen2.5-VL (3B)
2026.04
3.38
42
ε-Sampling
Backbone=Qwen2.5-VL (3B)
2026.04
3.36
35
Min-p
Backbone=Qwen2.5-VL (3B)
2026.04
3.32
43
Feedback
Search any
task
Search any
task