# Early evidence of how LLMs outperform traditional systems on OCR/HTR tasks for historical records

arXiv: https://arxiv.org/abs/2501.11623 · HTML: https://mlsanity.com/paper/2501.11623/

## Key facts

- Cited by 10 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 20 Jan 2025.

## Built on by (most cited first, 10 of 10)

- [Multimodal LLMs for OCR, OCR Post-Correction, and Named Entity Recognition in Historical Documents](https://mlsanity.com/paper/2504.00414/) (2025)
- [Benchmarking Large Language Models for Handwritten Text Recognition](https://mlsanity.com/paper/2503.15195/) (2025)
- [Evaluating LLMs for Historical Document OCR: A Methodological Framework for Digital Humanities](https://mlsanity.com/paper/2510.06743/) (2025)
- [Entropy Heat-Mapping: Localizing GPT-Based OCR Errors with Sliding-Window Shannon Analysis](https://mlsanity.com/paper/2505.00746/) (2025)
- [An HTR-LLM Workflow for High-Accuracy Transcription and Analysis of Abbreviated Latin Court Hand](https://mlsanity.com/paper/2507.04132/) (2025)
- [Improving MLLM Historical Record Extraction with Test-Time Image](https://mlsanity.com/paper/2509.09722/) (2025)
- [When Harmful Content Gets Camouflaged: Unveiling Perception Failure of LVLMs with CamHarmTI](https://mlsanity.com/paper/2512.03087/) (2025)
- [Quid est VERITAS? A Modular Framework for Archival Document Analysis](https://mlsanity.com/paper/2603.28108/) (2026)
- [ICDAR 2026 HIPE-OCRepair Competition on LLM-Assisted OCR Post-Correction for Historical Documents](https://mlsanity.com/paper/2607.08143/) (2026)
- [When Do VLMs Help Arabic Manuscript OCR? A Cross-Dataset Study](https://mlsanity.com/paper/2608.22366/) (2026)

## Abstract (arXiv)

We explore the ability of two LLMs -- GPT-4o and Claude Sonnet 3.5 -- to transcribe historical handwritten documents in a tabular format and compare their performance to traditional OCR/HTR systems: EasyOCR, Keras, Pytesseract, and TrOCR. Considering the tabular form of the data, two types of experiments are executed: one where the images are split line by line and the other where the entire scan is used as input. Based on CER and BLEU, we demonstrate that LLMs outperform the conventional OCR/HTR methods. Moreover, we also compare the evaluated CER and BLEU scores to human evaluations to better judge the outputs of whole-scan experiments and understand influential factors for CER and BLEU. Combining judgments from all the evaluation metrics, we conclude that two-shot GPT-4o for line-by-line images and two-shot Claude Sonnet 3.5 for whole-scan images yield the transcriptions of the historical records most similar to the ground truth.