---
title: "ATHAAR: A Medieval Scientific Arabic Manuscript Dataset for Handwritten Text Recognition"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2611.08091v1"
permalink: "https://feed.astralor.com/items/itm_5982ef00dca9f9e6c9345277"
published: "2026-11-05T14:36:16.000Z"
score: 57
channel: "ai"
tags: ["多模态","ATHAAR","Medieval Scientific Arabic","Manuscript Dataset","Handwritten Text Recognition","Despite","Arabic","VLMs"]
---

# ATHAAR: A Medieval Scientific Arabic Manuscript Dataset for Handwritten Text Recognition

## At a glance

Despite recent progress in handwritten text recognition, medieval Arabic scientific manuscripts remain an open challenge for current vision-language models (VLMs) due to cursive ligatures, variable diacritization, and m…

## Key points

- Despite recent progress in handwritten text recognition, medieval Arabic scientific manuscripts remain an open challenge for current vision-language models (VLMs) due to cursive ligatures, variable diacritization, and m…
- This paper aims to address this gap, by proposing the first annotation guidelines dedicated to the transcription of scientific Arabic manuscripts inspired by textual criticism, a well-established approach to historical …
- We then propose, ATHAAR, a new dataset manually annotated following this guideline.

## Full text

Despite recent progress in handwritten text recognition, medieval Arabic scientific manuscripts remain an open challenge for current vision-language models (VLMs) due to cursive ligatures, variable diacritization, and more importantly the lack of training data. This paper aims to address this gap, by proposing the first annotation guidelines dedicated to the transcription of scientific Arabic manuscripts inspired by textual criticism, a well-established approach to historical text editing. We then propose, ATHAAR, a new dataset manually annotated following this guideline. Unlike previous resources in the field, ATHAAR varies both witnesses of the same text, calligraphic styles, and content to deal with texts but also numerals. Relying on state-of-the-art VLMs, we finally design a set of experiments to study models transfer across scripts, chapters, and treatises. Our findings highlight the limitations of current systems when faced with the specificities of ancient scientific texts.

---

Verify against the original before citing; curated by Astral Feed: https://feed.astralor.com/items/itm_5982ef00dca9f9e6c9345277