Document Information Extraction on VRDU Registration Form Mixed Template
92.78Micro-F1LMDXPaLM 2-S
Evaluation Results
| Method | Links | |
|---|---|---|
| LMDXPaLM 2-S|D| (Training Data Size)=200, Modality=T+L, Localized?=true2023.09 | 92.78 | |
| LMDXPaLM 2-S|D| (Training Data Size)=100, Modality=T+L, Localized?=true2023.09 | 92.41 | |
| LMDXPaLM 2-S|D| (Training Data Size)=50, Modality=T+L, Localized?=true2023.09 | 91.42 | |
| FormNet|D| (Training Data Size)=200, Modality=T+L, Localized?=true2023.09 | 90.51 | |
| LayoutLMv3|D| (Training Data Size)=200, Modality=T+L+I, Localized?=true2023.09 | 89.77 | |
| LayoutLMv2|D| (Training Data Size)=200, Modality=T+L+I, Localized?=true2023.09 | 89.19 | |
| LayoutLMv2|D| (Training Data Size)=100, Modality=T+L+I, Localized?=true2023.09 | 88.36 | |
| FormNet|D| (Training Data Size)=100, Modality=T+L, Localized?=true2023.09 | 88.13 | |
| LayoutLM|D| (Training Data Size)=200, Modality=T+L+I, Localized?=true2023.09 | 87.94 | |
| LMDXPaLM 2-S|D| (Training Data Size)=10, Modality=T+L, Localized?=true2023.09 | 87.72 | |
| LayoutLMv3|D| (Training Data Size)=100, Modality=T+L+I, Localized?=true2023.09 | 87.32 | |
| LayoutLM|D| (Training Data Size)=100, Modality=T+L+I, Localized?=true2023.09 | 86.02 | |
| FormNet|D| (Training Data Size)=50, Modality=T+L, Localized?=true2023.09 | 85.38 | |
| LayoutLMv2|D| (Training Data Size)=50, Modality=T+L+I, Localized?=true2023.09 | 84.13 | |
| LayoutLMv3|D| (Training Data Size)=50, Modality=T+L+I, Localized?=true2023.09 | 81.36 | |
| LayoutLM|D| (Training Data Size)=50, Modality=T+L+I, Localized?=true2023.09 | 80.15 | |
| LMDXGemini Pro|D| (Training Data Size)=0, Modality=T+L, Localized?=true2023.09 | 75.15 | |
| LMDXPaLM 2-S|D| (Training Data Size)=0, Modality=T+L, Localized?=true2023.09 | 71.65 | |
| LayoutLMv2|D| (Training Data Size)=10, Modality=T+L+I, Localized?=true2023.09 | 69.44 | |
| Gemini Pro+OCR|D| (Training Data Size)=0, Modality=T, Localized?=false2023.09 | 69.41 | |
| GPT-4V+Image|D| (Training Data Size)=0, Modality=I, Localized?=false2023.09 | 65.34 | |
| GPT-3.5+OCR|D| (Training Data Size)=0, Modality=T, Localized?=false2023.09 | 63.86 | |
| FormNet|D| (Training Data Size)=10, Modality=T+L, Localized?=true2023.09 | 63.61 | |
| LayoutLMv3|D| (Training Data Size)=10, Modality=T+L+I, Localized?=true2023.09 | 60.72 | |
| PaLM 2-S+OCR|D| (Training Data Size)=0, Modality=T, Localized?=false2023.09 | 59.78 | |
| Gemini Pro+Image|D| (Training Data Size)=0, Modality=I, Localized?=false2023.09 | 48.6 | |
| LayoutLM|D| (Training Data Size)=10, Modality=T+L+I, Localized?=true2023.09 | 36.41 | |
| LLaVA-v1.5-13B+Image|D| (Training Data Size)=0, Modality=I, Localized?=false2023.09 | 5 |