Image-to-Web Conversion on IW-BENCH (Simple split)
69.4Element AccuracyGPT4V
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| GPT4VChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 69.4 | 66.4 | |
| WebSightChain-of-Thought (MCoT) status=Disabled2024.09 | 64.7 | 64.3 | |
| Qwen-VL-MaxChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 62.5 | 61.7 | |
| Gemini ProChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 61.7 | 60.8 | |
| Qwen-VL-MaxChain-of-Thought (MCoT) status=Disabled2024.09 | 59.8 | 58.8 | |
| Claude 3 OpusChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 57.3 | 56.7 | |
| Gemini ProChain-of-Thought (MCoT) status=Disabled2024.09 | 56.5 | 54.9 | |
| Claude 3 OpusChain-of-Thought (MCoT) status=Disabled2024.09 | 51.5 | 49.6 | |
| GPT4VChain-of-Thought (MCoT) status=Disabled2024.09 | 46.3 | 44.6 | |
| Qwen-VL-PlusChain-of-Thought (MCoT) status=Disabled2024.09 | 44.4 | 43.6 | |
| Qwen-VL-ChatChain-of-Thought (MCoT) status=Disabled2024.09 | 43.8 | 42.4 | |
| LLAVA-LLAMA-2-13BChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 40.1 | 38.8 | |
| LLaVA-LLaMA-2-13BChain-of-Thought (MCoT) status=Disabled2024.09 | 32.1 | 31.8 | |
| MPLUG-OWL2Chain-of-Thought (MCoT) status=Disabled2024.09 | 24.7 | 23.8 | |
| miniGPT-4-LLaMA-2-7BChain-of-Thought (MCoT) status=Disabled2024.09 | 20.5 | 19.4 | |
| LLaMA-Adapter-V2-7BChain-of-Thought (MCoT) status=Disabled2024.09 | 8.6 | 7.8 |