Detecting Latin in Historical Books with Large Language Models: A Multimodal Benchmark

Yu Wu, Ke Shu, Jonas Fischer, Lidia Pivovarova, David Rosson +2 more
10/22/2025
cs.CLcs.AIcs.CV

Official code: COMHIS/EACL26-detect-latin (2 stars, MIT, last commit 2026-03-25).

Abstract

This paper presents a novel task of extracting low-resourced and noisy Latin fragments from mixed-language historical documents with varied layouts. We benchmark and evaluate the performance of large foundation models against a multimodal dataset of 724 annotated pages. The results demonstrate that reliable Latin detection with contemporary zero-shot models is achievable, yet these models lack a functional comprehension of Latin. This study establishes a comprehensive baseline for processing Latin within mixed-language corpora, supporting quantitative analysis in intellectual history and historical linguistics. Both the dataset and code are available at https://github.com/COMHIS/EACL26-detect-latin.

View on arXivView PDF

Code Implementations(1)

COMHIS/EACL26-detect-latinOfficial100%
MIT

Official implementation and dataset of EACL paper 816: Detecting Latin in Historical Books with Large Language Models: A Multimodal Benchmark.

21PythonJan 21, 20267 months agoMIT

Similar Papers

LiMe: a Latin Corpus of Late Medieval Criminal Sentences

83% match2024

"Don't Teach Minerva": Guiding LLMs Through Complex Syntax for Faithful Latin Translation with RAG

81% match2025

Modern Models, Medieval Texts: A POS Tagging Study of Old Occitan

80% match2025

Loci Similes: A Benchmark for Extracting Intertextualities in Latin Literature

79% match2026

Adapting Multilingual Embedding Models to Historical Luxembourgish

79% match2025

LITERA: An LLM Based Approach to Latin-to-English Translation

79% match2025

Cite this paper

@article{wu2025detecting,
  title  = {Detecting Latin in Historical Books with Large Language Models: A Multimodal Benchmark},
  author = {Yu Wu and Ke Shu and Jonas Fischer and Lidia Pivovarova and David Rosson and Eetu Mäkelä and Mikko Tolonen},
  year   = {2025},
  eprint = {2510.19585},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2510.19585v3}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2510.19585.svg)](https://codeofpaper.com/paper/2510.19585)

Discussion