Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Understanding on LLaVA-Bench In-the-Wild
Loading...
5.833
Accuracy
ILVAD
5.209
5.371
5.533
5.695
May 20, 2026
Accuracy
Detailedness
Naturalness
Updated 13d ago
Evaluation Results
Method
Method
Links
Accuracy
Detailedness
Naturalness
ILVAD
Backbone=LLaVA-1.5-7B,...
2026.05
5.833
6.133
6.867
ILVAD
Backbone=Qwen2-VL-7B,...
2026.05
5.558
5.94
6.197
ILVAD
Backbone=LLaVA-NeXT-7B...
2026.05
5.467
6.633
7.633
LLaVA-1.5-7B
Decoding=Greedy
2026.05
5.45
5.833
7.017
LLaVA-NeXT-7B
Decoding=Greedy
2026.05
5.4
6.417
7.767
Qwen2-VL-7B
Decoding=Greedy
2026.05
5.233
5.992
6.11
Feedback
Search any
task
Search any
task