Our new X account is live! Follow @wizwand_team for updates
Home
/
Benchmarks
Humor Classification on MMSD Humor style 2.0 (test)
Loading...
80.71
Accuracy
SFT on Combined → GRPO (Combined)
60.7732
65.9491
71.125
76.3009
Jan 23, 2026
Accuracy
F1 Score
Updated 4d ago
Evaluation Results
Method
Method
Links
Accuracy
F1 Score
SFT on Combined → GRPO (Combined)
Prompting=zero-shot, S...
2026.01
80.71
93
Gemini 2.5 Flash
Prompting=zero-shot
2026.01
79.13
89
LLaMA 90B Vision Instruct
Prompting=zero-shot
2026.01
78.71
87
SFT on Combined → GRPO (Style specific)
Prompting=zero-shot, S...
2026.01
78.59
91
Qwen2.5-VL 32B Instruct
Prompting=zero-shot
2026.01
77.14
87
Phi-4 Multimodal
Prompting=zero-shot
2026.01
70.33
73
Gemma 3 27B
Prompting=zero-shot
2026.01
62.23
68
LLaVA 1.5 7B
Prompting=zero-shot
2026.01
61.54
61
Feedback
Search any
task
Search any
task