Share your thoughts, 1 month free Claude Pro on us
See more
Home
/
Benchmarks
Multimodal Emotion Reasoning on MEP-Bench
Loading...
80.17
Recall
HumanOmni-V2‡
47.8988
56.2769
64.655
73.0331
Jun 24, 2026
Recall
Accuracy @ V
Accuracy @ A
Updated 1mo ago
Evaluation Results
Method
Method
Links
Recall
Accuracy @ V
Accuracy @ A
HumanOmni-V2‡
Training Strategy=Orac...
2026.06
80.17
48.4
37.6
OPPO
Training Strategy=Ours
2026.06
70.44
76.4
50.6
Baseline
Training Strategy=Ours
2026.06
58
59.2
33
AffectGPT-R1
Training Strategy=RL B...
2026.06
54.85
54.2
30
Qwen2.5-Omni
Training Strategy=SFT...
2026.06
53
57
33.8
AffectGPT
Training Strategy=SFT...
2026.06
49.14
35
26
Feedback
Search any
task
Search any
task