Using 20 scanned pages of 1921 Belgian handwritten tables, the authors report that GPT-4o and Claude Sonnet 3.5 transcribe more accurately than EasyOCR, Keras, Pytesseract, and TrOCR, with two-shot prompting giving the best results.
International Journal on Document Analysis and Recognitio n 5(1), 39–46 (2002)
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
dataset 1
citation-polarity summary
fields
cs.CV 1years
2025 1verdicts
CONDITIONAL 1roles
dataset 1polarities
use dataset 1representative citing papers
citing papers explorer
-
Early evidence of how LLMs outperform traditional systems on OCR/HTR tasks for historical records
Using 20 scanned pages of 1921 Belgian handwritten tables, the authors report that GPT-4o and Claude Sonnet 3.5 transcribe more accurately than EasyOCR, Keras, Pytesseract, and TrOCR, with two-shot prompting giving the best results.