Funny Caption Generation on Humor in AI (#Top10)
66.41Top-1 AccuracyHOMER
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| HOMERBase LLM=GPT-4o2026.02 | 66.41 | 83.7 | 89.18 | |
| HOMERBase LLM=Claude-42026.02 | 64.67 | 82.67 | 87 | |
| HumorousAIBase LLM=Claude-42026.02 | 62.86 | 78.86 | 82.67 | |
| HumorousAIBase LLM=GPT-4o2026.02 | 62.11 | 81.24 | 85.15 | |
| Self-consistencyBase LLM=GPT-4o2026.02 | 62.03 | 77.96 | 82.93 | |
| Few-shotBase LLM=Claude-42026.02 | 61.67 | 78.7 | 82.67 | |
| CLoTBase LLM=GPT-4o2026.02 | 61.17 | 75.29 | 80 | |
| Self-consistencyBase LLM=Claude-42026.02 | 60.73 | 76.9 | 81.66 | |
| Few-shotBase LLM=GPT-4o2026.02 | 58.07 | 78.91 | 82.44 | |
| LoLBase LLM=Claude-42026.02 | 58.06 | 75.19 | 80 | |
| LoLBase LLM=GPT-4o2026.02 | 56.3 | 75.21 | 81.01 | |
| CoTBase LLM=GPT-4o2026.02 | 45.79 | 70.59 | 79.61 | |
| CLoTBase LLM=Claude-42026.02 | 43.15 | 53.25 | 56 | |
| CoTBase LLM=Claude-42026.02 | 37.31 | 47.62 | 51.01 | |
| HOMERBase LLM=Qwen-VL (7B)2026.02 | 24.06 | 41.75 | 49.59 | |
| CLoTBase LLM=Qwen-VL (7B)2026.02 | 21.67 | 36.67 | 43.33 | |
| Few-shotBase LLM=Qwen-VL (7B)2026.02 | 19.6 | 29.59 | 33.67 | |
| HumorousAIBase LLM=Qwen-VL (7B)2026.02 | 18.26 | 27.53 | 30.67 | |
| CoTBase LLM=Qwen-VL (7B)2026.02 | 16.76 | 27.33 | 33.01 | |
| PhunnyBase LLM=GPT-4o2026.02 | 16.09 | 27.47 | 32.94 | |
| Self-consistencyBase LLM=Qwen-VL (7B)2026.02 | 15.86 | 24.99 | 28.67 | |
| LoLBase LLM=Qwen-VL (7B)2026.02 | 15.12 | 23.84 | 28.17 | |
| PhunnyBase LLM=Claude-42026.02 | 14.24 | 32.87 | 46.4 | |
| HOMERBase LLM=LLaVA-1.5 (7B)2026.02 | 10.22 | 15.22 | 19.56 | |
| Few-shotBase LLM=LLaVA-1.5 (7B)2026.02 | 8.44 | 10.89 | 12.22 | |
| CLoTBase LLM=LLaVA-1.5 (7B)2026.02 | 6.4 | 8 | 8 | |
| PhunnyBase LLM=Qwen-VL (7B)2026.02 | 5.92 | 9.25 | 11.11 | |
| Self-consistencyBase LLM=LLaVA-1.5 (7B)2026.02 | 5.78 | 6.22 | 6.67 | |
| HumorousAIBase LLM=LLaVA-1.5 (7B)2026.02 | 4 | 10 | 13.33 | |
| PhunnyBase LLM=LLaVA-1.5 (7B)2026.02 | 3.89 | 10 | 13.33 | |
| LoLBase LLM=LLaVA-1.5 (7B)2026.02 | 1.33 | 4.09 | 6.67 | |
| CoTBase LLM=LLaVA-1.5 (7B)2026.02 | 1.11 | 1.11 | 1.11 |