Multimodal Humor Explanation on MemeCap, New Yorker, and YesBut Average
56.7Avg F1BOTTLEHUMOR
Evaluation results
| Method | Links | |
|---|---|---|
| BOTTLEHUMORModel=GPT4o2025.02 | 56.7 | |
| SR-noCModel=GPT4o2025.02 | 53.6 | |
| ZSModel=GPT4o2025.02 | 53.3 | |
| SRModel=GPT4o2025.02 | 52.2 | |
| COTModel=GPT4o2025.02 | 41.7 | |
| BOTTLEHUMORModel=Flash1.52025.02 | 37.6 | |
| SRModel=Flash1.52025.02 | 35.7 | |
| ZSModel=Flash1.52025.02 | 35.5 | |
| SR-noCModel=Flash1.52025.02 | 34.3 | |
| BOTTLEHUMORModel=Qwen22025.02 | 32.6 | |
| SR-noCModel=Qwen22025.02 | 32 | |
| ZSModel=Qwen22025.02 | 31.9 | |
| SRModel=Qwen22025.02 | 31.9 | |
| BOTTLEHUMORModel=Phi2025.02 | 26.2 | |
| COTModel=Flash1.52025.02 | 25.4 | |
| COTModel=Qwen22025.02 | 25.1 | |
| SRModel=Phi2025.02 | 23.4 | |
| SR-noCModel=Phi2025.02 | 23 | |
| ZSModel=Phi2025.02 | 22.7 | |
| COTModel=Phi2025.02 | 17.9 |