Image-to-Web Conversion on IW-BENCH Average
48.9Element AccuracyWebSight
Evaluation results
| Method | Links | ||
|---|---|---|---|
| WebSightChain-of-Thought (MCoT) status=Disabled2024.09 | 48.9 | 47.9 | |
| Qwen-VL-MaxChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 46.5 | 45.4 | |
| GPT4VChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 45.8 | 44.5 | |
| Gemini ProChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 43.2 | 42.5 | |
| Qwen-VL-MaxChain-of-Thought (MCoT) status=Disabled2024.09 | 39.9 | 38.5 | |
| Claude 3 OpusChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 39.3 | 38.7 | |
| Gemini ProChain-of-Thought (MCoT) status=Disabled2024.09 | 37.1 | 36.2 | |
| Claude 3 OpusChain-of-Thought (MCoT) status=Disabled2024.09 | 33.3 | 31.8 | |
| GPT4VChain-of-Thought (MCoT) status=Disabled2024.09 | 30.4 | 29.4 | |
| Qwen-VL-PlusChain-of-Thought (MCoT) status=Disabled2024.09 | 28.6 | 27.7 | |
| LLAVA-LLAMA-2-13BChain-of-Thought (MCoT) status=Enabled, MCoT Hops=52024.09 | 27.8 | 26.7 | |
| Qwen-VL-ChatChain-of-Thought (MCoT) status=Disabled2024.09 | 27.4 | 26.5 | |
| LLaVA-LLaMA-2-13BChain-of-Thought (MCoT) status=Disabled2024.09 | 20.3 | 19.7 | |
| MPLUG-OWL2Chain-of-Thought (MCoT) status=Disabled2024.09 | 16.1 | 14.9 | |
| miniGPT-4-LLaMA-2-7BChain-of-Thought (MCoT) status=Disabled2024.09 | 13.8 | 13.1 | |
| LLaMA-Adapter-V2-7BChain-of-Thought (MCoT) status=Disabled2024.09 | 5.5 | 5.3 |