Visual Question Answering on BizDocs (held-out)
76.4VQA ScoreGPT-4+OCR
Evaluation Results
| Method | Links | |
|---|---|---|
| GPT-4+OCRSize=~1T, Setting=ZS2023.12 | 76.4 | |
| DocLLM-7BSize=7B, Setting=STDD2023.12 | 73.3 | |
| DocLLM-1BSize=1B, Setting=STDD2023.12 | 65.5 | |
| Llama2+OCRSize=7B, Setting=ZS2023.12 | 48.4 |