Elamur Research⌕
☰
home / papers / 2026 / InkSight: Towards AI-Aided Historical Manuscript Analysis

InkSight: Towards AI-Aided Historical Manuscript Analysis

Paper · 2026 · Andrey Sakhovskiy, Ivan Ulitin, Emilia Bojarskaja et al.

Summary

Studying medieval Arabic manuscripts at scale is slow. Transcription demands rare paleographic expertise, a single difficult page can take days, and few software tools support the work. The authors present InkSight, an open web system that covers the full path from page image to research question. It has three stages. First, Kraken segments each page into lines and a four-step geometric and content filter cleans up the results. Second, a LoRA-fine-tuned Qwen3-VL-8B model transcribes each line. Third, a RAG layer indexes the book, retrieves the passages relevant to a user's query and passes them to an LLM, so answers come with explicit textual evidence. The OCR model was trained on Muharaf, a sample of synthetic SARD, and MAS, a new in-house corpus from Uzbek archives. On the KITAB benchmark and MAS, proprietary models such as GPT-4o and Gemini-2.0 Flash did well on modern text but fell behind small fine-tuned open models on historical pages. This suggests that low-resource manuscript digitization can work without costly language-specific pretraining.

Key points

  • Training data:
    • Muharaf: 1,644 pages, 36,311 lines, 19th–21st c.
    • SARD: 843,622 synthetic images, ~690M words, 10 fonts; 100k sampled for training.
    • MAS: 1,023 pages, 11,841 lines, 12th–19th c., Naskh/Nastaliq/Taliq scripts, sourced from the Oriental Studies institute named after Abu Rayhan Biruni.
  • OCR result: On MAS, fine-tuned Qwen3-VL-8B reached a CER of 0.24, versus 0.34 for the Arabic-pretrained AIN-7B. Qwen2.5-VL-7B+sft performed comparably.
  • RAG setup: The same encoder handles retrieval and reranking, with top-k = 4 and a similarity cutoff. Answers are generated by Qwen3.5-397B-A17B via OpenRouter.
  • Case study: A 17th-century copy of Dala'il al-Khayrat (186 pages, 3,720 segmented lines) was queried for mentions of trade routes, with answers returned in seconds.
  • Availability: A demo runs at inksight.ru and the code is on GitHub (ds-hub-sochi/InkSight-tool). The modular design allows adaptation to other languages.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

 Institutions

🎓Skolkovo Institute of Science and TechnologySkolkovo · Russia5 works · 142 citations

Also on the byline: Perm State University.

 Fields, topics and concepts

📚 Shelves 6The field of science this work belongs to (OpenAlex fields).Computer Science · 16,667 worksArtificial Intelligence · 16,194 worksArts and Humanities · 16,039 worksLanguage and Linguistics · 15,891 worksComputer Vision and Pattern Recognition · 2,869 worksLiterature and Literary Theory · 519 works

Cite

Cite as Andrey Sakhovskiy, Ivan Ulitin, Emilia Bojarskaja, Vladimir Kokh, Ruslan Murtazin, Maxim Novopoltsev and Semen Budennyy. 2026. InkSight: Towards AI-Aided Historical Manuscript Analysis. Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations).

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.18653/v1/2026.eacl-demo.20

@inproceedings{sakhovskiy2026inksight,
  title = {{InkSight: Towards AI-Aided Historical Manuscript Analysis}},
  author = {Andrey Sakhovskiy, Ivan Ulitin, Emilia Bojarskaja, Vladimir Kokh, Ruslan Murtazin, Maxim Novopoltsev and Semen Budennyy},
  booktitle = {Proceedings of the 19th Conference of the European Chapter of the Association for Computational Linguistics (Volume 3: System Demonstrations)},
  year = {2026},
  doi = {10.18653/v1/2026.eacl-demo.20},
  url = {https://doi.org/10.18653/v1/2026.eacl-demo.20}
}

References

36 references, 19 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ahmed Abdelali, Hamdy Mubarak, Shammur Chowdhury, Maram Hasanain, Basel Mousi, Sabri Boughorbel, Samir Abdaljalil, Yassine El Kheir, Daniel Izham, Fahim Dalvi, Majd Hawasly, Nizi Nazar, Youssef Elshahawy, Ahmed Ali, Nadir Durrani, Natasa Milic-Frayling, and Firoj Alam. 2024. LAraBench: Benchmarking Arabic AI with large language models. In Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics (Volume 1: Long Papers), pages 487–520, St. Julian’s, Malta. Association for Computational Linguistics. ↗
  2. Raghad Al-Rasheed, Abdullah Al Muaddi, Hawra Aljasim, Rawan Al-Matham, Muneera Alhoshan, Asma Al Wazrah, and Abdulrahman AlOsaimy. 2025. Evaluating RAG pipelines for Arabic lexical information retrieval: A comparative study of embedding and generation models. In Proceedings of the 1st Workshop on NLP for Languages Using Arabic Script, pages 155–164, Abu Dhabi, UAE. Association for Computational Linguistics.
  3. Mikel Artetxe, Sebastian Ruder, and Dani Yogatama. 2020. On the cross-lingual transferability of monolingual representations. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 4623–4637, Online. Association for Computational Linguistics.
  4. Jinze Bai, Shuai Bai, Shusheng Yang, Shijie Wang, Sinan Tan, Peng Wang, Junyang Lin, Chang Zhou, and Jingren Zhou. 2023. Qwen-vl: A versatile vision-language model for understanding, localization, text reading, and beyond. Preprint, arXiv:2308.12966. ↗
  5. Ayan Kumar Bhunia, Shuvozit Ghose, Amandeep Kumar, Pinaki Nath Chowdhury, Aneeshan Sain, and Yi-Zhe Song. 2021. Metahtr: Towards writer-adaptive handwritten text recognition. In 2021 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR), pages 15825–15834. ↗
  6. Houcine Boubaker, Abdelkarim Elbaati, Najiba Tagougui, Haikal El Abed, Monji Kherallah, Volker Märgner, and Adel M. Alimi. 2021. Adab database. find
  7. Adrian Chan, Anupam Mijar, Mehreen Saeed, Chau-Wai Wong, and Akram Khater. 2024. Hatformer: Historic handwritten arabic text recognition with transformers. arXiv preprint arXiv:2410.02179. ↗
  8. Jonathan H. Clark, Eunsol Choi, Michael Collins, Dan Garrette, Tom Kwiatkowski, Vitaly Nikolaev, and Jennimaria Palomaki. 2020. TyDi QA: A benchmark for information-seeking question answering in typologically diverse languages. Transactions of the Association for Computational Linguistics, 8:454–470. ↗
  9. Vladimir I Levenshtein. 1966. Binary codes capable of correcting deletions, insertions and reversals. Soviet Physics Doklady, 10:707. find
  10. Jacob Devlin, Ming-Wei Chang, Kenton Lee, and Kristina Toutanova. 2019. BERT: Pre-training of deep bidirectional transformers for language understanding. pages 4171–4186, Minneapolis, Minnesota.
  11. Patrick Lewis, Barlas Oguz, Ruty Rinott, Sebastian Riedel, and Holger Schwenk. 2020a. MLQA: Evaluating cross-lingual extractive question answering. In Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, pages 7315– 7330, Online. Association for Computational Linguistics.
  12. Ishak Dolek and Atakan Kurt. 2025. Ottoman htr: Recognition of the ottoman riqa font using deep learning models. 2025 Innovations in Intelligent Systems and Applications Conference (ASYU), pages 1–6. find
  13. Mohammad Fasha, Bassam Hammo, Nadim Obeid, and Jabir Widian. 2020. A hybrid deep learning model for arabic text recognition. CoRR, abs/2009.01987. ↗
  14. Heba Hassan, Ahmed El-Mahdy, and Mohamed E. Hussein. 2021. Arabic scene text recognition in the deep learning era: Analysis on a novel dataset. IEEE Access, 9:107046–107058. ↗
  15. Ahmed Heakl, Sara Ghaboura, Omkar Thawakar, Fahad Shahbaz Khan, Hisham Cholakkal, Rao Muhammad Anwer, and Salman H. Khan. 2025a. AIN: the arabic inclusive large multimodal model. CoRR, abs/2502.00094. ↗
  16. Ahmed Heakl, Muhammad Abdullah Sohail, Mukul Ranjan, Rania Elbadry, Ghazi Shazan Ahmad, Mohamed El-Geish, Omar Maher, Zhiqiang Shen, Fahad Shahbaz Khan, and Salman Khan. 2025b. KITAB-bench: A comprehensive multi-domain benchmark for Arabic OCR and document understanding. In Findings of the Association for Computational Linguistics: ACL 2025, pages 22006–22024, Vienna, Austria. Association for Computational Linguistics. ↗
  17. Edward J. Hu, Yelong Shen, Phillip Wallis, Zeyuan Allen-Zhu, Yuanzhi Li, Shean Wang, Lu Wang, and Weizhu Chen. 2022. Lora: Low-rank adaptation of large language models. In The Tenth International Conference on Learning Representations, ICLR 2022, Virtual Event, April 25-29, 2022. OpenReview.net.
  18. JaidedAI. 2020. Easyocr. https://github.com/ JaidedAI/EasyOCR. GitHub repository. ↗
  19. Benjamin Kiessling. 2019. Kraken - A Universal Text Recognizer for the Humanities. In Digital Humanities 2019, Utrecht, Netherlands. find
  20. Vu Trong Kim, Michael Krumdick, Varshini Reddy, Franck Dernoncourt, and Viet Dac Lai. 2024. An analysis of multilingual FActScore. In Proceedings of the 2024 Conference on Empirical Methods in Natural Language Processing, pages 4309–4333, Miami, Florida, USA. Association for Computational Linguistics. ↗
  21. Y. Lecun, L. Bottou, Y. Bengio, and P. Haffner. 1998. Gradient-based learning applied to document recognition. Proceedings of the IEEE, 86(11):2278–2324. ↗
  22. Patrick Lewis, Ethan Perez, Aleksandra Piktus, Fabio Petroni, Vladimir Karpukhin, Naman Goyal, Heinrich Küttler, Mike Lewis, Wen-tau Yih, Tim Rocktäschel, Sebastian Riedel, and Douwe Kiela. 2020b. Retrieval-augmented generation for knowledge-intensive NLP tasks. In Advances in Neural Information Processing Systems 33: Annual Conference on Neural Information Processing Systems 2020, NeurIPS 2020, December 6-12, 2020, virtual.
  23. Minghao Li, Tengchao Lv, Jingye Chen, Lei Cui, Yijuan Lu, Dinei A. F. Florêncio, Cha Zhang, Zhoujun Li, and Furu Wei. 2023. Trocr: Transformer-based optical character recognition with pre-trained models. In Thirty-Seventh AAAI Conference on Artificial Intelligence, AAAI 2023, Thirty-Fifth Conference on Innovative Applications of Artificial Intelligence, IAAI 2023, Thirteenth Symposium on Educational Advances in Artificial Intelligence, EAAI 2023, Washington, DC, USA, February 7-14, 2023, pages 13094– 13102. AAAI Press. find
  24. Xin Liu, Lechen Zhang, Sheza Munir, Yiyang Gu, and Lu Wang. 2025. VeriFact: Enhancing long-form factuality evaluation with refined fact extraction and reference facts. In Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing, pages 17919–17936, Suzhou, China. Association for Computational Linguistics.
  25. Sabri A. Mahmoud, Irfan Ahmad, Wasfi G. Al-Khatib, Mohammad Alshayeb, Mohammad Tanvir Parvez, Volker Märgner, and Gernot A. Fink. 2014. Khatt: An open arabic offline handwritten text database. Pattern Recognition, 47(3):1096–1112. Handwriting Recognition and other PR Applications. ↗
  26. Sewon Min, Kalpesh Krishna, Xinxi Lyu, Mike Lewis, Wen-tau Yih, Pang Koh, Mohit Iyyer, Luke Zettlemoyer, and Hannaneh Hajishirzi. 2023. FActScore: Fine-grained atomic evaluation of factual precision in long form text generation. In Proceedings of the 2023 Conference on Empirical Methods in Natural Language Processing, pages 12076–12100, Singapore. Association for Computational Linguistics. ↗
  27. Hussein Mozannar, Elie Maamary, Karl El Hajal, and Hazem Hajj. 2019. Neural Arabic question answering. In Proceedings of the Fourth Arabic Natural Language Processing Workshop, pages 108–118, Florence, Italy. Association for Computational Linguistics.
  28. Omer Nacar, Yasser Al-Habashi, Serry Sibaee, Adel Ammar, and Wadii Boulila. 2025. Sard: A large-scale synthetic arabic ocr dataset for book-style text recognition. Preprint, arXiv:2505.24600. ↗
  29. Tri Nguyen, Mir Rosenberg, Xia Song, Jianfeng Gao, Saurabh Tiwary, Rangan Majumder, and Li Deng. 2016. MS MARCO: A human generated machine reading comprehension dataset. In Proceedings of the Workshop on Cognitive Computation: Integrating neural and symbolic approaches 2016 co-located with the 30th Annual Conference on Neural Information Processing Systems (NIPS 2016), Barcelona, Spain, December 9, 2016, volume 1773 of CEUR Workshop Proceedings. CEUR-WS.org.
  30. OpenAI, :, Aaron Hurst, Adam Lerer, Adam P. Goucher, Adam Perelman, Aditya Ramesh, Aidan Clark, AJ Ostrow, Akila Welihinda, Alan Hayes, Alec Radford, Aleksander M adry, Alex Baker-Whitcomb, Alex Beutel, Alex Borzunov, Alex Carney, Alex Chow, Alex Kirillov, and 401 others. 2024. Gpt-4o system card. Preprint, arXiv:2410.21276. ↗
  31. Peng Wang, Shuai Bai, Sinan Tan, Shijie Wang, Zhihao Fan, Jinze Bai, Keqin Chen, Xuejing Liu, Jialin Wang, Wenbin Ge, Yang Fan, Kai Dang, Mengfei Du, Xuancheng Ren, Rui Men, Dayiheng Liu, Chang Zhou, Jingren Zhou, and Junyang Lin. 2024. Qwen2-vl: Enhancing vision-language model’s perception of the world at any resolution. Preprint, arXiv:2409.12191. ↗
  32. Curtis Wigington, Chris Tensmeyer, Brian L. Davis, William A. Barrett, Brian L. Price, and Scott Cohen. 2018. Start, follow, read: End-to-end full-page handwriting recognition. In Computer Vision - ECCV 2018 - 15th European Conference, Munich, Germany, September 8-14, 2018, Proceedings, Part VI, volume 11210 of Lecture Notes in Computer Science, pages 372–388. Springer. find
  33. Mokhtari Younes and Yousfi Abdellah. 2015. Segmentation of arabic handwritten text to lines. Procedia Computer Science, 73:115–121. find
  34. Vikas Paruchuri and Datalab Team. 2025. Surya: A lightweight document ocr and analysis toolkit. https://github.com/VikParuchuri/surya. GitHub repository. ↗
  35. Mehreen Saeed, Adrian Chan, Anupam Mijar, Joseph Moukarzel, Georges Habchi, Carlos Younes, Amin Elias, Chau-Wai Wong, and Akram Khater. 2024a. Muharaf: Manuscripts of handwritten arabic dataset for cursive text recognition. In Advances in Neural Information Processing Systems 38: Annual Conference on Neural Information Processing Systems 2024, NeurIPS 2024, Vancouver, BC, Canada, December 10 - 15, 2024. find
  36. Mehreen Saeed, Adrian Chan, Anupam Mijar, Joseph Moukarzel, Georges Habchi, Carlos Younes, Amin Elias, Chau-Wai Wong, and Akram Khater. 2024b. Muharaf: Manuscripts of handwritten arabic dataset for cursive text recognition. A find

Updated

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/sakhovskiy-2026-inksight-ai-aided-historical-manuscript/