📊 Showing 5 results | 📏 Metric: F1
| Rank | Model | Paper | F1 | Date | Code |
|---|---|---|---|---|---|
| 1 | LayoutLMv2LARGE (Excluding OCR mismatch) | LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding | 97.81 | 2020-12-29 | 📦 huggingface/transformers 📦 PaddlePaddle/PaddleOCR 📦 microsoft/unilm |
| 2 | RORE (GeoLayoutLM) | Modeling Layout Reading Order as Ordering Relations for Visually-rich Document Understanding | 96.97 | 2024-09-29 | 📦 chongzhangFDU/ROOR |
| 3 | LayoutLMv2LARGE | LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding | 96.61 | 2020-12-29 | 📦 huggingface/transformers 📦 PaddlePaddle/PaddleOCR 📦 microsoft/unilm |
| 4 | LayoutLMv2BASE | LayoutLMv2: Multi-modal Pre-training for Visually-Rich Document Understanding | 96.25 | 2020-12-29 | 📦 huggingface/transformers 📦 PaddlePaddle/PaddleOCR 📦 microsoft/unilm |
| 5 | ChatGPT 3.5 SpatialFormat | LAPDoc: Layout-Aware Prompting for Documents | 77.00 | 2024-02-15 | - |