---
title: "A Dataset of Wolof Ajami Manuscripts for HTR — Yousuf 2026"
description: "Systems for recognizing Arabic-script text struggle with African languages that use Ajami, the adapted Arabic orthographies of West Africa."
authors: ["Oreen Yousuf", "Elhadji Djibril Diagne", "Christian Høgel", "Beata Megyesi", "Joakim Nivre"]
year: 2026
venue: "Proceedings of the Fifteenth Language Resources and Evaluation Conference"
source: "https://aclanthology.org/2026.lrec-1.253/"
doi: "10.63317/4pz98ojeeqpw"
licence: "CC BY-NC 4.0 (ELRA, LREC proceedings)"
licence_url: "https://creativecommons.org/licenses/by-nc/4.0/"
summary_licence: "Elamur Cosmic License"
canonical: "https://research.elamur.ai/free/en/papers/yousuf-2026-dataset-wolof-ajami-manuscripts-htr/"
---
[Elamur Research](https://md.research.elamur.ai/free/en/) · [this page in HTML](https://research.elamur.ai/free/en/papers/yousuf-2026-dataset-wolof-ajami-manuscripts-htr/)

[home](https://md.research.elamur.ai/free/en/) / [papers](https://md.research.elamur.ai/free/en/papers/) / [2026](https://md.research.elamur.ai/free/en/papers/2026/page/3/) / A Dataset of Wolof Ajami Manuscripts for HTR and OCR

# A Dataset of Wolof Ajami Manuscripts for HTR and OCR

**Authors:** Oreen Yousuf, Elhadji Djibril Diagne, Christian Høgel, [Beata Megyesi](https://md.research.elamur.ai/free/en/authors/beata-megyesi/), [Joakim Nivre](https://md.research.elamur.ai/free/en/authors/joakim-nivre/) · **Year:** 2026 · **Venue:** Proceedings of the Fifteenth Language Resources and Evaluation Conference · **Licence:** [CC BY-NC 4.0 (ELRA, LREC proceedings)](https://creativecommons.org/licenses/by-nc/4.0/)

**Links:** [Publisher's page](https://aclanthology.org/2026.lrec-1.253/) · [PDF](https://aclanthology.org/2026.lrec-1.253.pdf) · [DOI 10.63317/4pz98ojeeqpw](https://doi.org/10.63317/4pz98ojeeqpw) · [OpenAlex](https://openalex.org/W7160193388)

**Citations:** 0 citations (OpenAlex, 2026-10-14)

## Summary
Systems for recognizing Arabic-script text struggle with African languages that use Ajami, the adapted Arabic orthographies of West Africa. Their training data lacks the extra letters, the dense vocalization and gemination diacritics (tashkīl) and the regional calligraphy found in these texts. The authors address this gap for Wolof by building the first dataset of Wolof Ajami manuscripts with manual segmentation and transcription. They took twentieth-century handwritten and printed documents from Touba, Senegal, held in Boston University's Ajami collections. In eScriptorium, a Wolof Ajami philologist annotated region and line polygons, reading order and diplomatic transcriptions, and a separate reviewer then corrected errors. The data is distributed in ALTO format. The authors also release a Keyman keyboard for the historical orthography. They tested seven OpenITI models built for historical Arabic-script documents, which produced CERs between about 62% and 82%. Errors centred on diacritics, letters specific to Wolof and confusion between dotted and undotted letters. The resource provides a benchmark for adapting HTR/OCR to Ajami and digitized text for NLP and historical linguistics.

## Key points
- The corpus has 19 manuscripts (the conclusion says 17): 313 pages, 4,743 lines, 8 handwritten and 11 printed, all in Maghrebi script. It contains 9 prose and 10 poetry texts, ranging from 31 to 1,151 lines.
- The dataset ("Wolof Ajami Handwritten Text Recognition Dataset") is released on Zenodo (doi:10.5281/zenodo.15691685). It includes images, region and line segmentation, line order and transcriptions.
- The baselines are seven OpenITI models. MS Mellon Print scored best (CER 0.625) and Urdu Print worst (0.822).
- Generalized is the only model trained on Maghrebi script, yet it ranked only third (CER 0.671), behind MS Mellon Print and MS Pretrained.
- The open-source Wolof Ajami Keyman keyboard (GitHub: oyousuf/Ajami-Manuscript-Keyboards) and the model outputs are also released.

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

Fields: [Linguistics & languages](https://md.research.elamur.ai/free/en/shelves/linguistics/)

Topics: [Semitic languages](https://md.research.elamur.ai/free/en/topics/semitic-languages/)

## Cite

The paper itself, not this page. BibTeX from the library's record of the work.

```bibtex
@inproceedings{yousuf2026dataset,
  title = {{A Dataset of Wolof Ajami Manuscripts for HTR and OCR}},
  author = {Yousuf, Oreen and Diagne, Elhadji Djibril and Høgel, Christian and Megyesi, Beata and Nivre, Joakim},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/4pz98ojeeqpw},
  url = {https://doi.org/10.63317/4pz98ojeeqpw}
}
```

- [← previous A Dataset of Psychiatric Hospital Notes with Temporal Information Annotations](https://md.research.elamur.ai/free/en/papers/miller-2026-dataset-psychiatric-hospital-notes-temporal/)
- [next → A Dialogue System for Second Language Learning with Dynamic Adaptation to In-Dialogue Difficulty Changes](https://md.research.elamur.ai/free/en/papers/morioka-2026-dialogue-system-second-language-learning/)
- [↑ back to papers of 2026](https://md.research.elamur.ai/free/en/papers/2026/page/3/)

## References

31 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

1. Ife Adebara, Hawau Olamide Toyin, Nahom Tesfu Ghebremichael, AbdelRahim Elmadany, and Muhammad Abdul-Mageed. 2025. Where are we? evaluating llm performance on african languages. arXiv preprint arXiv:2502.19582. [find](https://scholar.google.com/scholar?q=Ife+Adebara%2C+Hawau+Olamide+Toyin%2C+Nahom+Tesfu+Ghebremichael%2C+AbdelRahim+Elmadany%2C+and+Muhammad+Abdul%2DMageed%2E+2025%2E+Where+are+we%3F+evaluating+llm+performance+on+african+languages%2E+arXiv+preprint+arXiv%3A2)
2. David Ifeoluwa Adelani, Jesujoba Oluwadara Alabi, Angela Fan, Julia Kreutzer, Xiaoyu Shen, Machel Reid, Dana Ruiter, Dietrich Klakow, Peter Nabende, Ernie Chang, et al. 2022a. A few thousand translations go a long way! leveraging pre-trained models for african news translation. arXiv preprint arXiv:… [find](https://scholar.google.com/scholar?q=David+Ifeoluwa+Adelani%2C+Jesujoba+Oluwadara+Alabi%2C+Angela+Fan%2C+Julia+Kreutzer%2C+Xiaoyu+Shen%2C+Machel+Reid%2C+Dana+Ruiter%2C+Dietrich+Klakow%2C+Peter+Nabende%2C+Ernie+Chang%2C+et+al%2E+2022a%2E+A+few+thousand+translati) [our card](https://md.research.elamur.ai/free/en/papers/adelani-2022-few-thousand-translations-go-long/)
3. David Ifeoluwa Adelani, Graham Neubig, Sebastian Ruder, Shruti Rijhwani, Michael Beukman, Chester Palen-Michel, Constantine Lignos, Jesujoba O Alabi, Shamsuddeen H Muhammad, Peter Nabende, et al. 2022b. Masakhaner 2.0: Africa-centric transfer learning for named entity recognition. arXiv preprint arX… [find](https://scholar.google.com/scholar?q=David+Ifeoluwa+Adelani%2C+Graham+Neubig%2C+Sebastian+Ruder%2C+Shruti+Rijhwani%2C+Michael+Beukman%2C+Chester+Palen%2DMichel%2C+Constantine+Lignos%2C+Jesujoba+O+Alabi%2C+Shamsuddeen+H+Muhammad%2C+Peter+Nabende%2C+et+al%2E+2022) [our card](https://md.research.elamur.ai/free/en/papers/adelani-2022-masakhaner-2-0-africa-centric/)
4. Birhanu Belay, Tewodros Habtegebrial, Million Meshesha, Marcus Liwicki, Gebeyehu Belay, and Didier Stricker. 2020. Amharic ocr: an end-to-end learning. Applied Sciences, 10(3):1117. [find](https://scholar.google.com/scholar?q=Birhanu+Belay%2C+Tewodros+Habtegebrial%2C+Million+Meshesha%2C+Marcus+Liwicki%2C+Gebeyehu+Belay%2C+and+Didier+Stricker%2E+2020%2E+Amharic+ocr%3A+an+end%2Dto%2Dend+learning%2E+Applied+Sciences%2C+10%283%29%3A1117%2E)
5. Birhanu Hailu Belay, Isabelle Guyon, Tadele Mengiste, Bezawork Tilahun, Marcus Liwicki, Tesfa Tegegne, and Romain Egele. 2024. A historical handwritten dataset for ethiopic ocr with baseline models and human-level performance. In International Conference on Document Analysis and Recognition, pages 2… [find](https://scholar.google.com/scholar?q=Birhanu+Hailu+Belay%2C+Isabelle+Guyon%2C+Tadele+Mengiste%2C+Bezawork+Tilahun%2C+Marcus+Liwicki%2C+Tesfa+Tegegne%2C+and+Romain+Egele%2E+2024%2E+A+historical+handwritten+dataset+for+ethiopic+ocr+with+baseline+models+an)
6. Emanuela Boroş, Ahmed Hamdi, Elvys Linhares Pontes, Luis-Adrián Cabrera-Diego, Jose G Moreno, Nicolas Sidere, and Antoine Doucet. 2020. Alleviating digitization errors in named entity recognition for historical documents. In Proceedings of the 24th conference on computational natural language learni… [find](https://scholar.google.com/scholar?q=Emanuela+Boro%C5%9F%2C+Ahmed+Hamdi%2C+Elvys+Linhares+Pontes%2C+Luis%2DAdri%C3%A1n+Cabrera%2DDiego%2C+Jose+G+Moreno%2C+Nicolas+Sidere%2C+and+Antoine+Doucet%2E+2020%2E+Alleviating+digitization+errors+in+named+entity+recognition+for+)
7. Marta R Costa-Jussà, James Cross, Onur Çelebi, Maha Elbayad, Kenneth Heafield, Kevin Heffernan, Elahe Kalbassi, Janice Lam, Daniel Licht, Jean Maillard, et al. 2022. No language left behind: Scaling human-centered machine translation. arXiv preprint arXiv:2207.04672. [find](https://scholar.google.com/scholar?q=Marta+R+Costa%2DJuss%C3%A0%2C+James+Cross%2C+Onur+%C3%87elebi%2C+Maha+Elbayad%2C+Kenneth+Heafield%2C+Kevin+Heffernan%2C+Elahe+Kalbassi%2C+Janice+Lam%2C+Daniel+Licht%2C+Jean+Maillard%2C+et+al%2E+2022%2E+No+language+left+behind%3A+Scaling+h)
8. Cheikh M Bamba Dione, David Adelani, Peter Nabende, Jesujoba Alabi, Thapelo Sindane, Happy Buzaaba, Shamsuddeen Hassan Muhammad, Chris Chinenye Emezue, Perez Ogayo, Anuoluwapo Aremu, et al. 2023. Masakhapos: Part-of-speech tagging for typologically diverse african languages. arXiv preprint arXiv:230… [find](https://scholar.google.com/scholar?q=Cheikh+M+Bamba+Dione%2C+David+Adelani%2C+Peter+Nabende%2C+Jesujoba+Alabi%2C+Thapelo+Sindane%2C+Happy+Buzaaba%2C+Shamsuddeen+Hassan+Muhammad%2C+Chris+Chinenye+Emezue%2C+Perez+Ogayo%2C+Anuoluwapo+Aremu%2C+et+al%2E+2023%2E+Masa)
9. Bonaventure FP Dossou, Atnafu Lambebo Tonja, Oreen Yousuf, Salomey Osei, Abigail Oppong, Iyanuoluwa Shode, Oluwabusayo Olufunke Awoyomi, and Chris Chinenye Emezue. 2022. Afrolm: A self-active learning-based multilingual pretrained language model for 23 african languages. arXiv preprint arXiv:2211.03… [find](https://scholar.google.com/scholar?q=Bonaventure+FP+Dossou%2C+Atnafu+Lambebo+Tonja%2C+Oreen+Yousuf%2C+Salomey+Osei%2C+Abigail+Oppong%2C+Iyanuoluwa+Shode%2C+Oluwabusayo+Olufunke+Awoyomi%2C+and+Chris+Chinenye+Emezue%2E+2022%2E+Afrolm%3A+A+self%2Dactive+learning)
10. Safiullah Faizullah, Muhammad Sohaib Ayub, Sajid Hussain, and Muhammad Asad Khan. 2023. A survey of ocr in arabic language: applications, techniques, and challenges. Applied Sciences, 13(7):4584. [find](https://scholar.google.com/scholar?q=Safiullah+Faizullah%2C+Muhammad+Sohaib+Ayub%2C+Sajid+Hussain%2C+and+Muhammad+Asad+Khan%2E+2023%2E+A+survey+of+ocr+in+arabic+language%3A+applications%2C+techniques%2C+and+challenges%2E+Applied+Sciences%2C+13%287%29%3A4584%2E)
11. Oana Ignat, Jean Maillard, Vishrav Chaudhary, and Francisco Guzmán. 2022. Ocr improves machine translation for low-resource languages. arXiv preprint arXiv:2202.13274. [find](https://scholar.google.com/scholar?q=Oana+Ignat%2C+Jean+Maillard%2C+Vishrav+Chaudhary%2C+and+Francisco+Guzm%C3%A1n%2E+2022%2E+Ocr+improves+machine+translation+for+low%2Dresource+languages%2E+arXiv+preprint+arXiv%3A2202%2E13274%2E)
12. Mahmoud SalahEldin Kasem, Mohamed Mahmoud, and Hyun-Soo Kang. 2023. Advancements and challenges in arabic optical character recognition: A comprehensive survey. ACM Computing Surveys. [find](https://scholar.google.com/scholar?q=Mahmoud+SalahEldin+Kasem%2C+Mohamed+Mahmoud%2C+and+Hyun%2DSoo+Kang%2E+2023%2E+Advancements+and+challenges+in+arabic+optical+character+recognition%3A+A+comprehensive+survey%2E+ACM+Computing+Surveys%2E)
13. Benjamin Kiessling, Gennady Kurin, Matthew Thomas Miller, and Kader Smail. 2024. Advances and limitations in open source arabic-script ocr: A case study. arXiv preprint arXiv:2402.10943. — [arXiv:2402.10943](https://arxiv.org/abs/2402.10943) [find](https://scholar.google.com/scholar?q=Benjamin+Kiessling%2C+Gennady+Kurin%2C+Matthew+Thomas+Miller%2C+and+Kader+Smail%2E+2024%2E+Advances+and+limitations+in+open+source+arabic%2Dscript+ocr%3A+A+case+study%2E+arXiv+preprint+arXiv%3A2402%2E10943%2E+%E2%80%94+%5BarXiv%3A2402)
14. Benjamin Kiessling, Robin Tissot, Peter Stokes, and Daniel Stökl Ben Ezra. 2019. escriptorium: an open source platform for historical document analysis. In 2019 international conference on document analysis and recognition workshops (icdarw), volume 2, pages 19–19. IEEE. [find](https://scholar.google.com/scholar?q=Benjamin+Kiessling%2C+Robin+Tissot%2C+Peter+Stokes%2C+and+Daniel+St%C3%B6kl+Ben+Ezra%2E+2019%2E+escriptorium%3A+an+open+source+platform+for+historical+document+analysis%2E+In+2019+international+conference+on+document+an)
15. Daniel Lopresti. 2008. Optical character recognition errors and their effects on natural language processing. In Proceedings of the second workshop on Analytics for Noisy Unstructured Text Data, pages 9–16. [find](https://scholar.google.com/scholar?q=Daniel+Lopresti%2E+2008%2E+Optical+character+recognition+errors+and+their+effects+on+natural+language+processing%2E+In+Proceedings+of+the+second+workshop+on+Analytics+for+Noisy+Unstructured+Text+Data%2C+pages)
16. Mohammed Lutf, Xinge You, Yiu-ming Cheung, and CL Philip Chen. 2014. Arabic font recognition based on diacritics features. Pattern Recognition, 47(2):672–684. [find](https://scholar.google.com/scholar?q=Mohammed+Lutf%2C+Xinge+You%2C+Yiu%2Dming+Cheung%2C+and+CL+Philip+Chen%2E+2014%2E+Arabic+font+recognition+based+on+diacritics+features%2E+Pattern+Recognition%2C+47%282%29%3A672%E2%80%93684%2E)
17. Matthew Thomas Miller, Maxim G Romanov, and Sarah Bowen Savant. 2018. Digitizing the textual heritage of the premodern islamicate world: Principles and plans. International Journal of Middle East Studies, 50(1):103–109. [find](https://scholar.google.com/scholar?q=Matthew+Thomas+Miller%2C+Maxim+G+Romanov%2C+and+Sarah+Bowen+Savant%2E+2018%2E+Digitizing+the+textual+heritage+of+the+premodern+islamicate+world%3A+Principles+and+plans%2E+International+Journal+of+Middle+East+Stud)
18. Meikal Mumin. 2014. The arabic script in africa: understudied literacy. In The Arabic Script in Africa, pages 41–76. Brill. [find](https://scholar.google.com/scholar?q=Meikal+Mumin%2E+2014%2E+The+arabic+script+in+africa%3A+understudied+literacy%2E+In+The+Arabic+Script+in+Africa%2C+pages+41%E2%80%9376%2E+Brill%2E)
19. Fallou Ngom. 2016. Muslims beyond the Arab world: The odyssey of ajami and the Muridiyya. Oxford University Press. [find](https://scholar.google.com/scholar?q=Fallou+Ngom%2E+2016%2E+Muslims+beyond+the+Arab+world%3A+The+odyssey+of+ajami+and+the+Muridiyya%2E+Oxford+University+Press%2E)
20. Fallou Ngom. 2017. West african manuscripts in arabic and african languages and digital preservation. In Oxford research encyclopedia of African history. [find](https://scholar.google.com/scholar?q=Fallou+Ngom%2E+2017%2E+West+african+manuscripts+in+arabic+and+african+languages+and+digital+preservation%2E+In+Oxford+research+encyclopedia+of+African+history%2E)
21. Fallou Ngom, Daivi Rodima-Taylor, and David Robinson. 2023. ajamı̄ literacies of africa: The hausa, fula, mandinka, and wolof traditions. Islamic Africa, 14(2):119–143. [find](https://scholar.google.com/scholar?q=Fallou+Ngom%2C+Daivi+Rodima%2DTaylor%2C+and+David+Robinson%2E+2023%2E+ajam%C4%B1%CC%84+literacies+of+africa%3A+The+hausa%2C+fula%2C+mandinka%2C+and+wolof+traditions%2E+Islamic+Africa%2C+14%282%29%3A119%E2%80%93143%2E)
22. Odunayo Ogundepo, Tajuddeen R Gwadabe, Clara E Rivera, Jonathan H Clark, Sebastian Ruder, David Ifeoluwa Adelani, Bonaventure FP Dossou, Abdou Aziz Diop, Claytone Sikasote, Gilles Hacheme, et al. 2023. Afriqa: Cross-lingual open-retrieval question answering for african languages. arXiv preprint arXi… [find](https://scholar.google.com/scholar?q=Odunayo+Ogundepo%2C+Tajuddeen+R+Gwadabe%2C+Clara+E+Rivera%2C+Jonathan+H+Clark%2C+Sebastian+Ruder%2C+David+Ifeoluwa+Adelani%2C+Bonaventure+FP+Dossou%2C+Abdou+Aziz+Diop%2C+Claytone+Sikasote%2C+Gilles+Hacheme%2C+et+al%2E+2023)
23. Olalekan Joseph Oni and Franklin Oladiipo Asahiah. 2020. Computational modelling of an optical character recognition system for yorùbá printed text images. Scientific African, 9:e00415. [find](https://scholar.google.com/scholar?q=Olalekan+Joseph+Oni+and+Franklin+Oladiipo+Asahiah%2E+2020%2E+Computational+modelling+of+an+optical+character+recognition+system+for+yor%C3%B9b%C3%A1+printed+text+images%2E+Scientific+African%2C+9%3Ae00415%2E)
24. Maxim Romanov, Matthew Thomas Miller, Sarah Bowen Savant, and Masoumeh Seydi. 2019. Open islamicate texts initiative: A machine-readable corpus of texts produced in the premodern islamicate world (poster). In Digital Humanities 2019 Conference Papers (9-12 July 2019). [find](https://scholar.google.com/scholar?q=Maxim+Romanov%2C+Matthew+Thomas+Miller%2C+Sarah+Bowen+Savant%2C+and+Masoumeh+Seydi%2E+2019%2E+Open+islamicate+texts+initiative%3A+A+machine%2Dreadable+corpus+of+texts+produced+in+the+premodern+islamicate+world+%28pos)
25. Utrecht University, 2019. [find](https://scholar.google.com/scholar?q=Utrecht+University%2C+2019%2E)
26. Eddie Antonio Santos and Atticus Harrigan. 2020. Design and evaluation of a smartphone keyboard for plains cree syllabics. In Proceedings of the 1st Joint Workshop on Spoken Language Technologies for Under-Resourced Languages (SLTU) and Collaboration and Computing for Under-Resourced Languages (CCUR… [find](https://scholar.google.com/scholar?q=Eddie+Antonio+Santos+and+Atticus+Harrigan%2E+2020%2E+Design+and+evaluation+of+a+smartphone+keyboard+for+plains+cree+syllabics%2E+In+Proceedings+of+the+1st+Joint+Workshop+on+Spoken+Language+Technologies+for+)
27. Esther H Showalter. 2016. Mobile device keyboard customization for a newly constructed orthography of a rural west african language. In Proceedings of the Eighth International Conference on Information and Communication Technologies and Development, pages 1–4. [find](https://scholar.google.com/scholar?q=Esther+H+Showalter%2E+2016%2E+Mobile+device+keyboard+customization+for+a+newly+constructed+orthography+of+a+rural+west+african+language%2E+In+Proceedings+of+the+Eighth+International+Conference+on+Informatio)
28. Daniel Van Strien, Kaspar Beelen, Mariona Coll Ardanuy, Kasra Hosseini, Barbara McGillivray, and Giovanni Colavizza. 2020. Assessing the impact of ocr quality on downstream nlp tasks. [find](https://scholar.google.com/scholar?q=Daniel+Van+Strien%2C+Kaspar+Beelen%2C+Mariona+Coll+Ardanuy%2C+Kasra+Hosseini%2C+Barbara+McGillivray%2C+and+Giovanni+Colavizza%2E+2020%2E+Assessing+the+impact+of+ocr+quality+on+downstream+nlp+tasks%2E)
29. Oreen Yousuf, Abdulmalik Aminu, Musa Salih Muhammad, Bashir Usman, Mustapha Kurfi Hashim, Joakim Nivre, Beáta Megyesi, and Christian Høgel. 2025. A handwritten text recognition dataset for ajami manuscripts in fulfulde and hausa. In International Conference on Document Analysis and Recognition, page… [find](https://scholar.google.com/scholar?q=Oreen+Yousuf%2C+Abdulmalik+Aminu%2C+Musa+Salih+Muhammad%2C+Bashir+Usman%2C+Mustapha+Kurfi+Hashim%2C+Joakim+Nivre%2C+Be%C3%A1ta+Megyesi%2C+and+Christian+H%C3%B8gel%2E+2025%2E+A+handwritten+text+recognition+dataset+for+ajami+manus)
30. Language Resource References Yousuf, Oreen. 2026. Wolof Ajami Keyboard. Keyman keyboard to transcribe Wolof Ajami manuscripts. The keyboard is designed specifically to transcribe the Wolof Ajami manuscripts sourced from Boston University used in this paper: [https://github.com/oyousuf/Ajami-Manuscrip…](https://github.com/oyousuf/Ajami-Manuscrip…) [↗](https://github.com/oyousuf/Ajami-Manuscrip…)
31. Yousuf, Oreen and Diagne, Elhadji Djibril and Nivre, Joakim and Megyesi, Beáta and Høgel, Christian. 2026. Wolof Ajami Handwritten Text Recognition Dataset. Handwritten and Printed Wolof Ajami manuscripts for HTR and OCR: https://doi.org/10.5281/zenodo.15691685. — [doi:10.5281/zenodo.15691685](https://doi.org/10.5281/zenodo.15691685) [↗](https://doi.org/10.5281/zenodo.15691685)