mlsanityData to 7 Oct 2026

A Free Lunch? Adapting PP-OCRv6 for Historical Text Recognition

arXiv 2609.20064 ↗First posted 17 Sep 2026cs.CVcited by 0

On the map: Multimodal LLMs / vision-language models → Document understanding and multimodal RAG

Abstract (arXiv)

Despite impressive reported scores, large vision-language models have seen limited practical uptake in historical automatic text recognition because of their computational cost, dependence on large-scale pretraining, and hallucination. Historical ATR therefore continues to rely largely on compact CRNN line recognizers, which are visually grounded and trainable on modest data. Lightweight recurrence-free recognizers promise the accuracy of larger models with the practical advantages of CRNNs, yet have not been comprehensively evaluated on historical writing. We adapt PP-OCRv6, a recent compact text recognizer without strong language modeling, for historical line recognition and compare it with a conventional CRNN across generalized pretraining, domain-specific training, corpus-level fine-tuning, and manuscript-specific few-shot adaptation on multilingual Latin- and Arabic-script material. While PP-OCRv6 does not consistently outperform the baseline when trained from scratch, heterogeneous pretraining produces markedly better generalization. Comparisons with the Qwen3.5-based Medusa recognizer further show that fine-tuned PP-OCRv6 can outperform a large VLM tailored towards historical Latin-script HTR.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in Document understanding and multimodal RAG, one of 16 directions in Multimodal LLMs / vision-language models. Builds on 19 of its 36 references that are in the map. First posted on arXiv 17 Sep 2026.

Builds on

19 of 36 references.

Closest ancestors · 4

Same direction, newest first

  1. 2026PP-OCRv6: From 1.5M to 34.5M Parameters, Surpassing Billion-Scale VLMs on OCR Tasks11 Jun
  2. 2026Reading or Guessing? Visual Grounding Failures of Vision-Language Models for OCR in Ancient Greek Editions26 May
  3. 2026PP-OCRv5: A Specialized 5M-Parameter Model Rivaling Billion-Parameter Vision-Language Models on OCR Tasks25 Mar
  4. 2026LightOnOCR: A 1B End-to-End Multilingual Vision-Language Model for State-of-the-Art OCR20 Jan

Other references · 15

  1. 2023Qwen-VL: A Versatile Vision-Language Model for Understanding, Localization, Text Reading, and Beyondcited by 2,179
  2. 2015An End-to-End Trainable Neural Network for Image-based Sequence Recognition and Its Application to Scene Text Recognitioncited by 519
  3. 2021TrOCR: Transformer-based Optical Character Recognition with Pre-trained Modelscited by 278
  4. 2018NRTR: A No-Recurrence Sequence-to-Sequence Model For Scene Text Recognitioncited by 73
  5. 2022DAN: a Segmentation-free Document Attention Network for Handwritten Document Recognitioncited by 62
  6. 2020OrigamiNet: Weakly-Supervised, Segmentation-Free, One-Step, Full Page Text Recognition by learning to unfoldcited by 53
  7. 2023DTrOCR: Decoder-only Transformer for Optical Character Recognitioncited by 40
  8. 2018Accurate, Data-Efficient, Unconstrained Text Recognition with Convolutional Neural Networkscited by 31
  9. 2024HTR-VT: Handwritten Text Recognition with Vision Transformercited by 24
  10. 2024SVTRv2: CTC Beats Encoder-Decoder Models in Scene Text Recognitioncited by 23
  11. 2018Calamari - A High-Performance Tensorflow-based Deep Learning Package for Optical Character Recognitioncited by 15
  12. 2019Fully Convolutional Networks for Handwriting Recognitioncited by 10
3 more
  1. 2024On the Generalization of Handwritten Text Recognition Modelscited by 10
  2. 2025CHURRO: Making History Readable with an Open-Weight Large Vision-Language Model for High-Accuracy, Low-Cost Historical Text Recognitioncited by 9
  3. 2022Easter2.0: Improving convolutional models for handwritten text recognitioncited by 6

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2026Combining Synthetic and Real Data for Low-Resource Historical OCR: A Manchu Case Study
  2. 2025Finetuning Vision-Language Models as OCR Systems for Low-Resource Languages: A Case Study of Manchu
  3. 2026A Benchmark of State-Space Models vs. Transformers and BiLSTM-based Models for Historical Newspaper OCR
  4. 2025Multimodal LLMs for OCR, OCR Post-Correction, and Named Entity Recognition in Historical Documents
  5. 2026When Simpler Is Better: Evaluating Translation Pipelines for Medieval Latin Manuscripts
  6. 2026Error Patterns in Historical OCR: A Comparative Analysis of TrOCR and a Vision-Language Model
  7. 2026Structure-Aware Text Recognition for Ancient Greek Critical Editions
  8. 2026Scaling State-Space Models from Lines to Paragraphs: An Ablation of Mamba-based OCR