Visual Information Extraction on CORD LayoutLLM (test)
85.53ANLS (%)Qwen2-VL-7B+
Evaluation Results
| Method | Links | |
|---|---|---|
| Qwen2-VL-7B+zero-shot=true, RIDGE fine-tuned=true2025.04 | 85.53 | |
| Qwen2-VL-7Bzero-shot=true, RIDGE fine-tuned=false2025.04 | 80.4 | |
| Monkeyzero-shot=true, RIDGE fine-tuned=false2025.04 | 69.48 | |
| LLaVA-NeXT-Mistral-7B+zero-shot=true, RIDGE fine-tuned=true2025.04 | 63.93 | |
| LLaVA-NeXT-Mistral-7Bzero-shot=true, RIDGE fine-tuned=false2025.04 | 63.81 | |
| DocOwl-1.5-Chatzero-shot=true, RIDGE fine-tuned=false2025.04 | 63.15 | |
| LayoutLLMzero-shot=true, RIDGE fine-tuned=false2025.04 | 63.1 |