Advances in handwritten text recognition have enabled large-scale transcription of historical documents, but still provide limited access to interpretable visual measurements for paleography, the study of historical scripts. In this paper, our main insight is that morphological script analysis, in particular the capacity to learn character prototypes from line-level transcriptions, enables the definition of scalable, meaningful, and stable paleographic measurements. More precisely, we leverage a transformer-based detection architecture together with a prototype-based line reconstruction module to learn prototypical characters and their occurrence, deformation, and positioning.<br>Our contributions are twofold. First, we introduce a deep architecture and learning methodology that enables efficient character modeling with only line-level transcription supervision, significantly improving over the Learnable Typewriter baseline and enabling accurate character bounding box prediction, unlocking its potential for paleographic measurements. Second, we introduce and demonstrate the paleographical relevance of automatic measurements enabled by our architecture for characters, bi-grams, and spaces between graphical units. For this demonstration, we extend the annotations of the codex Paris, BnF, fr. 2813, commissioned in the late fourteenth century by Charles V and copied by four hands, to 160 pages. We visualize our measurements over these pages, showing how they enable us not only to differentiate graphical profiles, but also to discover and analyze subtle variations. This case study outlines the scalability of our approach and its frugality in terms of required training data, since a single column of text is sufficient to compute our measurements on each of the 160 pages.</p>\n","updatedAt":"2026-06-12T08:13:10.617Z","author":{"_id":"66c9a126dc312dec43f34f98","avatarUrl":"/avatars/814f7ebd9a3fd626be898e92afa7e3fa.svg","fullname":"Raphael","name":"RaphaelBfr","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.8903177380561829},"editors":["RaphaelBfr"],"editorAvatarUrls":["/avatars/814f7ebd9a3fd626be898e92afa7e3fa.svg"],"reactions":[],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.09446","authors":[{"_id":"6a2a7443eb926374a3821968","name":"Malamatenia Vlachou Efstathiou","hidden":false},{"_id":"6a2a7443eb926374a3821969","user":{"_id":"66c9a126dc312dec43f34f98","avatarUrl":"/avatars/814f7ebd9a3fd626be898e92afa7e3fa.svg","isPro":false,"fullname":"Raphael","user":"RaphaelBfr","type":"user","name":"RaphaelBfr"},"name":"Raphaël Baena","status":"claimed_verified","statusLastChangedAt":"2026-06-12T07:42:52.413Z","hidden":false},{"_id":"6a2a7443eb926374a382196a","name":"Dominique Stutzmann","hidden":false},{"_id":"6a2a7443eb926374a382196b","name":"Mathieu Aubry","hidden":false}],"publishedAt":"2026-06-08T00:00:00.000Z","submittedOnDailyAt":"2026-06-12T00:00:00.000Z","title":"Leveraging Morphology for Historical Script Metrological Analysis","submittedOnDailyBy":{"_id":"66c9a126dc312dec43f34f98","avatarUrl":"/avatars/814f7ebd9a3fd626be898e92afa7e3fa.svg","isPro":false,"fullname":"Raphael","user":"RaphaelBfr","type":"user","name":"RaphaelBfr"},"summary":"Advances in handwritten text recognition have enabled large-scale transcription of historical documents, but still provide limited access to interpretable visual measurements for paleography, the study of historical scripts. In this paper, our main insight is that morphological script analysis, in particular the capacity to learn character prototypes from line-level transcriptions, enables the definition of scalable, meaningful, and stable paleographic measurements. More precisely, we leverage a transformer-based detection architecture together with a prototype-based line reconstruction module to learn prototypical characters and their occurrence, deformation, and positioning.\n Our contributions are twofold. First, we introduce a deep architecture and learning methodology that enables efficient character modeling with only line-level transcription supervision, significantly improving over the Learnable Typewriter baseline and enabling accurate character bounding box prediction, unlocking its potential for paleographic measurements. Second, we introduce and demonstrate the paleographical relevance of automatic measurements enabled by our architecture for characters, bi-grams, and spaces between graphical units. For this demonstration, we extend the annotations of the codex Paris, BnF, fr. 2813, commissioned in the late fourteenth century by Charles V and copied by four hands, to 160 pages. We visualize our measurements over these pages, showing how they enable us not only to differentiate graphical profiles, but also to discover and analyze subtle variations. This case study outlines the scalability of our approach and its frugality in terms of required training data, since a single column of text is sufficient to compute our measurements on each of the 160 pages.\n Data and code are publicly available at: https://malamatenia.github.io/morphology4metrology-analysis.","upvotes":0,"discussionId":"6a2a7443eb926374a382196c","projectPage":"https://malamatenia.github.io/morphology4metrology-analysis/","githubRepo":"https://github.com/raphael-baena/morphology4metrology","githubRepoAddedBy":"user","ai_summary":"A transformer-based architecture with prototype learning enables scalable paleographic measurements from historical documents using only line-level transcriptions, demonstrating its effectiveness on a 160-page codex with minimal training data requirements.","ai_keywords":["transformer-based detection architecture","prototype-based line reconstruction module","character prototypes","line-level transcription","learnable typewriter","character bounding box prediction"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","githubStars":12,"organization":{"_id":"65b7830478003ec6b3ac3571","name":"LesPonts","fullname":"Ecole des Ponts ParisTech","avatar":"https://www.gravatar.com/avatar/b346dcfe9d9e877e2d3f9fd64a6b7d66?d=retro&size=100"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[],"acceptLanguages":["en"],"organization":{"_id":"65b7830478003ec6b3ac3571","name":"LesPonts","fullname":"Ecole des Ponts ParisTech","avatar":"https://www.gravatar.com/avatar/b346dcfe9d9e877e2d3f9fd64a6b7d66?d=retro&size=100"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.09446.md","query":{}}">
Leveraging Morphology for Historical Script Metrological Analysis
Abstract
A transformer-based architecture with prototype learning enables scalable paleographic measurements from historical documents using only line-level transcriptions, demonstrating its effectiveness on a 160-page codex with minimal training data requirements.
Advances in handwritten text recognition have enabled large-scale transcription of historical documents, but still provide limited access to interpretable visual measurements for paleography, the study of historical scripts. In this paper, our main insight is that morphological script analysis, in particular the capacity to learn character prototypes from line-level transcriptions, enables the definition of scalable, meaningful, and stable paleographic measurements. More precisely, we leverage a transformer-based detection architecture together with a prototype-based line reconstruction module to learn prototypical characters and their occurrence, deformation, and positioning.
Our contributions are twofold. First, we introduce a deep architecture and learning methodology that enables efficient character modeling with only line-level transcription supervision, significantly improving over the Learnable Typewriter baseline and enabling accurate character bounding box prediction, unlocking its potential for paleographic measurements. Second, we introduce and demonstrate the paleographical relevance of automatic measurements enabled by our architecture for characters, bi-grams, and spaces between graphical units. For this demonstration, we extend the annotations of the codex Paris, BnF, fr. 2813, commissioned in the late fourteenth century by Charles V and copied by four hands, to 160 pages. We visualize our measurements over these pages, showing how they enable us not only to differentiate graphical profiles, but also to discover and analyze subtle variations. This case study outlines the scalability of our approach and its frugality in terms of required training data, since a single column of text is sufficient to compute our measurements on each of the 160 pages.
Data and code are publicly available at: https://malamatenia.github.io/morphology4metrology-analysis.
Community
Advances in handwritten text recognition have enabled large-scale transcription of historical documents, but still provide limited access to interpretable visual measurements for paleography, the study of historical scripts. In this paper, our main insight is that morphological script analysis, in particular the capacity to learn character prototypes from line-level transcriptions, enables the definition of scalable, meaningful, and stable paleographic measurements. More precisely, we leverage a transformer-based detection architecture together with a prototype-based line reconstruction module to learn prototypical characters and their occurrence, deformation, and positioning.
Our contributions are twofold. First, we introduce a deep architecture and learning methodology that enables efficient character modeling with only line-level transcription supervision, significantly improving over the Learnable Typewriter baseline and enabling accurate character bounding box prediction, unlocking its potential for paleographic measurements. Second, we introduce and demonstrate the paleographical relevance of automatic measurements enabled by our architecture for characters, bi-grams, and spaces between graphical units. For this demonstration, we extend the annotations of the codex Paris, BnF, fr. 2813, commissioned in the late fourteenth century by Charles V and copied by four hands, to 160 pages. We visualize our measurements over these pages, showing how they enable us not only to differentiate graphical profiles, but also to discover and analyze subtle variations. This case study outlines the scalability of our approach and its frugality in terms of required training data, since a single column of text is sufficient to compute our measurements on each of the 160 pages.
Upload images, audio, and videos by dragging in the text input, pasting, or clicking here.
Tap or paste here to upload images
Cite arxiv.org/abs/2606.09446 in a Space README.md to link it from this page.
Discussion (0)
Sign in to join the discussion. Free account, 30 seconds — email code or GitHub.
Sign in →No comments yet. Sign in and be the first to say something.