Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Vision-Audio-Language (VAL) on CMM
Loading...
84.5
PA Accuracy (Yes Instances)
DPO
79.3
80.65
82
83.35
Apr 15, 2026
PA Accuracy (Yes Instances)
HR Accuracy (No Instances)
Accuracy
Updated 3d ago
Evaluation Results
Method
Method
Links
PA Accuracy (Yes Instances)
HR Accuracy (No Instances)
Accuracy
DPO
2026.04
84.5
96.5
90.5
Base model
2026.04
83.5
96
89.8
SFT
2026.04
83.5
97
90.3
OmniDPO
2026.04
79.5
98.5
89
ACPO
2026.04
79.5
97
88.3
Feedback
Search any
task
Search any
task