Humorous Captioning on Culture-Aware Humorous Captioning Western Culture Context 1.0 (test)
7.32IRgemini-3-flash-preview
Evaluation Results
| Method | Links | |||||||
|---|---|---|---|---|---|---|---|---|
| gemini-3-flash-preview2026.04 | 7.32 | 5.71 | 5.86 | 8.14 | 6.92 | 6.25 | 6.7 | |
| GPT-4o2026.04 | 7.27 | 6.11 | 5.91 | 8.06 | 6.76 | 6.47 | 6.76 | |
| Claude Sonnet 4.5_pred2026.04 | 7.11 | 5.66 | 5.89 | 8.05 | 6.82 | 6.21 | 6.62 | |
| QwenVL+CLoT2026.04 | 7.11 | 4.32 | 4.38 | 7.8 | 5.67 | 5.02 | 5.72 | |
| InternVL3-8B-Instruct2026.04 | 7.08 | 4.9 | 4.95 | 7.79 | 6 | 5.61 | 6.06 | |
| Qwen3-VL-8B-Instruct2026.04 | 7.04 | 5.49 | 5.61 | 7.84 | 6.41 | 6.2 | 6.43 | |
| cogvlm2-llama3-chat-19B2026.04 | 6.92 | 4.93 | 5.03 | 7.76 | 5.8 | 5.57 | 6 | |
| CuHAlign2026.04 | 6.88 | 6.63 | 5.6 | 7.84 | 6.67 | 6.16 | 6.63 | |
| MiniCPM-V 2.62026.04 | 6.82 | 5.17 | 5.21 | 7.75 | 5.86 | 5.82 | 6.1 | |
| Qwen2.5-VL-7B-Instruct2026.04 | 6.6 | 4.94 | 4.93 | 7.61 | 5.92 | 5.55 | 5.92 | |
| MiniGPT-42026.04 | 6.54 | 3.94 | 2.95 | 6.98 | 3.97 | 3.36 | 4.62 | |
| LLaVA-OneVision-7B2026.04 | 6.02 | 4.67 | 4.44 | 6.97 | 4.55 | 4.64 | 5.21 | |
| GLM-4V-9B2026.04 | 4.79 | 3.7 | 3.6 | 5.67 | 2.25 | 3.01 | 3.84 |