Key Information Extraction on KLC (test)
60.3F1 ScoreDocLLM-7B
Evaluation results
| Method | Links | |
|---|---|---|
| DocLLM-7BModel Size=7B, Modality=Text+Layout, Evaluation Setting=SDDS2023.12 | 60.3 | |
| DocLLM-1BModel Size=1B, Modality=Text+Layout, Evaluation Setting=SDDS2023.12 | 58.9 | |
| GPT-4+OCRModel Size=~1T, Modality=Text, Evaluation Setting=Zero-Shot (ZS)2023.12 | 45.9 | |
| mPLUG-DocOwl 1.5backbone=mPLUG-Owl 2, fine-tuning dataset=4M2024.11 | 38.7 | |
| HVFAbackbone=mPLUG-Owl, fine-tuning dataset=650K2024.11 | 36.7 | |
| UReaderModel Size=~7B, Modality=Text+Vision, Evaluation Setting=SDDS2023.12 | 32.8 | |
| mPLUG-DocOwlModel Size=~7B, Modality=Text+Vision, Evaluation Setting=SDDS2023.12 | 30.3 | |
| Llama2+OCRModel Size=7B, Modality=Text, Evaluation Setting=Zero-Shot (ZS)2023.12 | 27.8 |