Funny caption generation on Electric sheep High-Humor
75.53Recall@1HOMER
Evaluation results
| Method | Links | |||
|---|---|---|---|---|
| HOMERBase LLM=GPT-4o2026.02 | 75.53 | 89.21 | 92.1 | |
| HumorousAIBase LLM=GPT-4o2026.02 | 67.39 | 80.57 | 83.38 | |
| CLoTBase LLM=GPT-4o2026.02 | 63.33 | 71.83 | 77.33 | |
| HOMERBase LLM=Claude-42026.02 | 62.27 | 81.37 | 86.94 | |
| LoLBase LLM=GPT-4o2026.02 | 61.26 | 79.22 | 84.55 | |
| LoLBase LLM=Claude-42026.02 | 60.6 | 79.33 | 83 | |
| HumorousAIBase LLM=Claude-42026.02 | 59.4 | 77.66 | 83.33 | |
| Self-consistencyBase LLM=Claude-42026.02 | 57.72 | 74.39 | 79.13 | |
| CoTBase LLM=GPT-4o2026.02 | 57.52 | 76.13 | 81.19 | |
| CoTBase LLM=Claude-42026.02 | 57.52 | 69 | 72.51 | |
| Few-shotBase LLM=GPT-4o2026.02 | 50.67 | 69.33 | 80.67 | |
| Self-consistencyBase LLM=GPT-4o2026.02 | 48.57 | 64.95 | 74.23 | |
| CLoTBase LLM=Claude-42026.02 | 41.67 | 62 | 68.33 | |
| Few-shotBase LLM=Claude-42026.02 | 32.67 | 54 | 66.67 | |
| PhunnyBase LLM=GPT-4o2026.02 | 26.22 | 36.13 | 45.92 | |
| HOMERBase LLM=Qwen-VL (7B)2026.02 | 22.74 | 36.18 | 41.58 | |
| PhunnyBase LLM=Claude-42026.02 | 20.16 | 38.58 | 48.33 | |
| Few-shotBase LLM=Qwen-VL (7B)2026.02 | 19.38 | 30.96 | 35.73 | |
| HOMERBase LLM=LLaVA-1.5 (7B)2026.02 | 19.33 | 23.17 | 25 | |
| CoTBase LLM=Qwen-VL (7B)2026.02 | 19.06 | 30.9 | 36.66 | |
| LoLBase LLM=Qwen-VL (7B)2026.02 | 17.58 | 28.53 | 31.5 | |
| Self-consistencyBase LLM=Qwen-VL (7B)2026.02 | 15.26 | 21.61 | 24.74 | |
| Few-shotBase LLM=LLaVA-1.5 (7B)2026.02 | 14 | 16 | 16.67 | |
| CLoTBase LLM=LLaVA-1.5 (7B)2026.02 | 13.33 | 16.11 | 16.67 | |
| HumorousAIBase LLM=Qwen-VL (7B)2026.02 | 11.8 | 16.83 | 18.99 | |
| HumorousAIBase LLM=LLaVA-1.5 (7B)2026.02 | 11.11 | 17.78 | 22.22 | |
| CLoTBase LLM=Qwen-VL (7B)2026.02 | 8 | 21.1 | 26.67 | |
| PhunnyBase LLM=Qwen-VL (7B)2026.02 | 4.44 | 13.33 | 22.22 | |
| PhunnyBase LLM=LLaVA-1.5 (7B)2026.02 | 4.17 | 13.33 | 22.36 | |
| LoLBase LLM=LLaVA-1.5 (7B)2026.02 | 1.9 | 5.71 | 9.52 | |
| Self-consistencyBase LLM=LLaVA-1.5 (7B)2026.02 | 1.37 | 3.99 | 6.67 | |
| CoTBase LLM=LLaVA-1.5 (7B)2026.02 | 1.08 | 3.66 | 5.56 |