Humor Detection on YesBut
70.4AccuracyQwen2.5-VL
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2.5-VL#Params=7B, CoT=true2026.01 | 70.4 | |
| Qwen2.5-VL#Params=7B, CoT=false2026.01 | 68.8 | |
| InternVL3#Params=8B, CoT=true2026.01 | 66.4 | |
| InternVL3#Params=8B, CoT=false2026.01 | 65.6 | |
| MoCoT (Ours)#Params=3B2026.01 | 62.9 | |
| Qwen2.5-VL#Params=3B, CoT=true2026.01 | 57.7 | |
| LLaVA-1.6#Params=7B, CoT=false2026.01 | 56.9 | |
| Phi-3.5#Params=4B, CoT=false2026.01 | 56.9 | |
| Qwen2.5-VL#Params=3B, CoT=false2026.01 | 55.7 | |
| LLaVA-1.6#Params=7B, CoT=true2026.01 | 54.9 | |
| Ovis2#Params=2B, CoT=false2026.01 | 53.8 | |
| Phi-3.5#Params=4B, CoT=true2026.01 | 51 | |
| Gemma-3#Params=4B, CoT=false2026.01 | 51 | |
| Ovis2#Params=2B, CoT=true2026.01 | 50.6 | |
| XComposer-2.5#Params=7B, CoT=false2026.01 | 50.2 | |
| Mono#Params=2B, CoT=false2026.01 | 48.2 | |
| InternVL2.5#Params=2B, CoT=true2026.01 | 48.2 | |
| Gemma-3#Params=4B, CoT=true2026.01 | 47 | |
| XComposer-2.5#Params=7B, CoT=true2026.01 | 45.5 | |
| InternVL2.5#Params=2B, CoT=false2026.01 | 45.5 | |
| Mono#Params=2B, CoT=true2026.01 | 32.8 |